BioMed CentralProteome Science

ss

Open AcceResearchSample prep for proteomics of breast cancer proteomics and gene ontology reveal dramatic differences in protein solubilization preferences of radioimmunoprecipitation assay and urea lysis buffersLambert CM Ngoka12

Address 1Department of Chemistry Virginia Commonwealth University 1001 West Main Street P O Box 842006 Richmond VA 23284-2006 USA and 2Department of Biomedical Sciences Paul L Foster School of Medicine Texas Tech University Health Sciences Center MSB-1 5001 El Paso Drive El Paso TX 79905 USA

Email Lambert CM Ngoka - lngokavcuedu

AbstractBackground An important step in the proteomics of solid tumors including breast cancer consists ofefficiently extracting most of proteins in the tumor specimen For this purpose Radio-Immunoprecipitation Assay (RIPA) buffer is widely employed RIPA buffers rapid and highly efficient celllysis and good solubilization of a wide range of proteins is further augmented by its compatibility withprotease and phosphatase inhibitors ability to minimize non-specific protein binding leading to a lowerbackground in immunoprecipitation and its suitability for protein quantitation

Results In this work the insoluble matter left after RIPA buffer extraction of proteins from breast tumorsare subjected to another extraction step using a urea-based buffer It is shown that RIPA and urea lysisbuffers fractionate breast tissue proteins primarily on the basis of molecular weights The averagemolecular weight of proteins that dissolve exclusively in urea buffer is up to 60 higher than in RIPA

Gene Ontology (GO) and Directed Acyclic Graphs (DAG) are used to map the collective biological andbiophysical attributes of the RIPA and urea proteomes The Cellular Component and Molecular Functionannotations reveal protein solubilization preferences of the buffers especially the compartmentalizationand functional distributions

It is shown that nearly all extracellular matrix proteins (ECM) in the breast tumors and matched normaltissues are found nearly exclusively in the urea fraction while they are mostly insoluble in RIPA bufferAdditionally it is demonstrated that cytoskeletal and extracellular region proteins are more soluble in ureathan in RIPA whereas for nuclear cytoplasmic and mitochondrial proteins RIPA buffer is preferred

Extracellular matrix proteins are highly implicated in cancer including their proteinase-mediateddegradation and remodelling tumor development progression adhesion and metastasis Thus if they arenot efficiently extracted by RIPA buffer important information may be missed in cancer research

Conclusion For proteomics of solid tumors a two-step extraction process is recommended Firstproteins in the tumor specimen should be extracted with RIPA buffer Second the RIPA-insoluble materialshould be extracted with the urea-based buffer employed in this work

Published 24 October 2008

Proteome Science 2008 630 doi1011861477-5956-6-30

Received 10 March 2008Accepted 24 October 2008

This article is available from httpwwwproteomescicomcontent6130

copy 2008 Ngoka licensee BioMed Central Ltd This is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby20) which permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Page 1 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

BackgroundOver the past few years proteomics has emerged as a pow-erful new technology capable of generating unprece-dented details of protein maps in a wide range of cell typesand disease processes Increasingly however it is becom-ing recognized that the success of a proteomic experimentis critically dependent on the sample preparation step Anideal sample prep protocol should not only isolate asmuch of the proteins of interest as possible from the bio-logical source but also preserve optimal sample integrityand morphology It should also present the entire samplein a form that is compatible with optimum mass spectro-metric analysis

Proteins in their native states are generally embedded intheir natural environments where they are associated withother proteins biological macromolecules or other matrixmaterials They may also be components of multi-proteincomplexes integrated into plasma membranes ororganelles They are generally insoluble in their nativestates once isolation from their biological environmentsThey must therefore be denatured in order to bring theminto solution This ultimately entails dissociating thechemical bonds connecting them in their native statesThe bonds and appropriate agentsmethods for dissociat-ing them [1] include disulfide bond (reduction amp alkyla-tion) hydrogen bond (chaotropes) electrostaticinteractions (salts charged detergents chaotropes) charge-dipole (chaotropes) dipole-dipole (strong dipolar mole-cules) van der Waals (salt dipolar molecules chaotropes)and hydrophobic interactions (salts dipolar molecules cha-otropes)

Effective sample preparation for proteomics disrupts theseassociations and solubilizes as large a subset of the pro-teins as possible Sample solubilization buffers typicallycontain a number of additives (chaotropes detergentsreducing agents buffers salts and ampholytes) In pro-teomics perhaps two of the most effective and widelyemployed lysis buffers for extracting proteins from cellsand tissues are Radio-Immunoprecipitation Assay buffer(RIPA buffer) [2] and urea lysis buffer [134]

The base ingredients of a typical RIPA buffer include 50mM Tris HCl pH 8 150 mM NaCl 1 NP-40 05sodium deoxycholate and 01 SDS Protease and phos-phatase inhibitors are additionally added prior to usedepending on the application (these are usually notadded when preparing lysates for phosphatase assays)Additional optimization of the lysis procedure or substi-tution of the base ingredients may be required for eachspecific application for example PBS pH 74 can substi-tute for both Tris HCl and NaCl An example variant ofRIPA buffer that contains protease and phosphataseinhibitors consists of 10 mM Tris pH 74 100 mM NaCl

1 mM EDTA 1 mM EGTA 1 mM NaF 20 mM Na4P2O7 2mM Na3VO4 01 SDS 05 sodium deoxycholate 1Triton-X 100 10 glycerol 1 mM PMSF (made from a 03M stock in DMSO) or 1 mM AEBSF (water soluble versionof PMSF) 60 μgmL aprotinin 10 μgmL leupeptin 1 μgmL pepstatin (alternatively protease inhibitor cocktailmay be used)

RIPA buffers rapid and efficient cell lysis and solubiliza-tion of a wide range of proteins including cytoplasmicmembrane and nuclear proteins makes it a standard forWestern blotting Its versatility is further augmented by itscompatibility with protease and phosphatase inhibitorsstability and ability to minimize non-specific protein-binding interactions leading to low backgrounds inimmunoprecipitation Still RIPA buffer is very compati-ble with a myriad of applications including reporterassays protein assays immunoassays and protein purifi-cation When protein quantitation is desired RIPA bufferis the lysis buffer of choice due to its compatibility withthe BCA Protein Assay although it can denature kinases[5] and can disrupt protein-protein interactions in immu-noprecipitationpull down assays [6]

Urea buffer is another versatile and efficient cell and tissuelysing buffer whose typical composition include TRISbase 40 mM Urea 7 M Thiourea 2 M NP-40 or CHAPS4 DTT 10 mM Urea is used at concentrations rangingfrom 5 to 9 M often with thiourea at concentrations up to2 M The additive thiourea can dramatically enhance thesolubility of a wide range of proteins ndash nuclear mem-brane cytosolic and including even tubulin that is highlyprone to aggregation in urea buffer [134] As in RIPAbuffer different detergents and buffers can be substitutedfor the buffering base NP-40 or CHAPS depending onthe application Urea inactivates proteases that degradecellular proteins [7] Therefore there is little need to addprotease inhibitors However urea and thiourea canhydrolyze to cyanate and thiocyanate respectively whichcan modify amino groups on proteins (eg carbamyla-tion of proteins by isocyanate) and this hydrolysis is pro-moted by heat

Extracellular matrix (ECM) proteins (collagens fibronec-tin vitronectin thrombospondin laminins tenascinosteopontin and entatin etc) are highly implicated intumor cell growth motility angiogenesis apoptosis pro-liferation invasion and metastasis [8-22] They contributeto signal transduction differentiation site-specific geneexpression immunological functions wound healing andinflammation [81517182223] They and their remod-eling enzymes including urokinase plasmogen activatorsmatrix metalloproteinases and cathepsins L B and D arealso known to profoundly modulate mammary glandbranching morphogenesis [24] They are also important

Page 2 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

biomarkers of breast cancer For example Veeck and co-workers [9] showed recently that the extracellular matrixprotein Inter-α-trypsin inhibitor is strongly down-regu-lated in all human breast cancers

Thus in cancer proteomics especially where it is desiredto recover as much of the cellular or tissue proteins as pos-sible it is important not to completely rely on a singlelysis buffer that could exclude an entire class of criticallyneeded proteins especially the low abundance proteinsIndeed there are numerous proteomics publications thatfall exactly under the above category In breast cancer pro-teomics for example articles are found (example Refs[2325-30]) wherein researchers wanted to extract all pro-teins present in the breast tumor specimens but usedRIPA buffer as the sole buffer

In a study of ECM components in differentiating terato-carcinoma cells Grover and Adamson [31] noted thatfibronectin was poorly solubilized in RIPA buffer Other-wise no other direct reference suggesting insolubility ofECMs in RIPA buffer was found as yet Instead researchers(examples Refs [913-152032]) routinely use RIPAbuffer as the sole lysis and solubilization buffer in theimmunoprecipitation [15] immunoblot [131420] and

Western blot [932] analysis of extracellular matrix pro-teins

MethodsThe key steps (Figure 1) include protein extraction frombreast tumors and matched normal breast tissues sampleclean-up with GE Healthcare tools trypsin digestiondesalting with Michrom cartridges mass spectrometry2Dnano-LCESI-MSMS database search and protein IDdata processing and bioinformatics

Breast tumor tissue lysatesProtein lysates of breast tumors and matching normalbreast tissues are obtained from Protein BiotechnologiesInc (Ramona CA) The samples are selected to representdifferent types of tumors (T2-018 Tumor T2-048 TumorT2-029 Tumor) non-neoplastic normal breast tissue (T2-048 Normal) Diagnosis Stage Grade Age and GrossFindings (Table 1)

Protein extractionLysates provided by Protein Biotechnologies wereextracted by a two-step procedure First proteins areextracted with a modified Radio-ImmunoprecipitationAssay (RIPA) lysis buffer to yield the soluble fraction Sec-

Proteomics Workflow flow chartFigure 1Proteomics Workflow flow chartoutlining the key experimental steps in this work

Page 3 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

ond the residual insoluble fraction left after RIPA bufferextraction is subjected to additional extraction using aurea-based buffer to produce a second protein fractionThe compositions of the lysis buffers are as follows

Extraction 1Radioimmunoprecipitation Assay Lysis Buffer (modi-fied) Base Ingredients (PBS pH 74 SDS 01 Na-deox-ycholate 025) RIPA Protease Inhibitors (1 mMPhenylmethylsulfonyl fluoride (PMSF) EDTA 1 mM (cal-cium chelator 100 mM stock solution in H2O pH 74)Leupeptin 1 μgmL Aprotinin 1 μgmL Pepstatin 1 μgmL NaF (1 mM)) RIPA Phosphatase Inhibitors (Acti-vated Na3VO4 (1 mM))

Extraction 2Urea Lysis Buffer (PBS pH 74 50 M Urea 20 M Thiou-rea 50 mM DTT 01 SDS)

Cleanup of lysatesIn the cleanup step the proteins are separated from buff-ers detergents salts and other contaminants using amethod that is largely derived from a protocol and 2Dclean-up kit provided by Amersham Biosciences (GEHealthcare Piscataway NJ) The kit consists of four rea-gents a precipitant that precipitates the proteins to formpellets a co-precipitant that co-precipitates with the pro-teins and enhances their removal from solution a washbuffer that removes non-protein contaminants from theprotein precipitate and a wash additive that promotesrapid and complete re-suspension of the sample proteins

Prior to the beginning of clean-up the wash buffer waschilled at -20degC for 1 hr After thawing and spinning

down 100 μg aliquots of breast tumor lysates andmatched normal breast tissue lysates 300 μL of the pre-cipitant was added The mixture was vortexed on Eppen-dorf Thermomixer R (Eppendorf North AmericaWestbury NY) and then incubated in ice for 15 minutesNext 300 μL of co-precipitant was added and the mixturemixed The mixture was centrifuged at 12000 times g for 5minutes to pellet the proteins The clear supernatant liq-uid was carefully pipetted out while retaining the proteinprecipitate at the bottom of the 15 mL Eppendorf tubeWithout disturbing the pellet layer 40 μL of co-precipi-tant was added to the supernatant through the tilted sideof the 15 mL Eppendorf tube The mixture was kept in icefor 5 minutes before centrifuging it again at 12000 times g foranother 5 minutes The pellet was dispersed by adding 25μL of MilliQ water and centrifuging for 10 minutes Afteradding 1 mL of chilled wash buffer at -20degC and 5 μL ofwash additive the mixture was vortexed once every 30 sec-onds for a total of 35 minutes At this point the proteinsdid not dissolve but dispersed The mixture was againcentrifuged at 12000 times g for 5 minutes The supernatantwas carefully discarded and the pellet dried The pelletsare amorphous

In-solution digestionThe dried pellet was re-suspended in 20 μL 8 M urea100mM ammonium bicarbonate (ABC) and 06 μL of 100mM Dithiothreitol (DTT) in 100 mM ABC (ie 3 mMDTT) was stirred in Eppendorf Thermomixer R (Eppen-dorf North America Westbury NY) for 1 hr at 29degC Afteradjusting to room temperature 15 μL of 200 mM iodoa-cetamide (IAA) in 100 mM ABC (final concentration of 15mM IAA) was added Alkylation was then carried out byincubating the mixture for 45 minutes in a darkroom

Table 1 Characteristics of the breast tumors and matched normal breast tissues analyzed in this work

T2-018 Tumor T2-048 Tumor T2-048 Normal T2-029 Tumor

Location Right breast Left breast Left breast Bilateral

Diagnosis Infiltrating ductal carcinoma Infiltrating ductal carcinoma Normal Adenocarcinoma

Stage IV T4N0M0 IIB T2N1M0 Normal IV T4N1M0

Grade II I I III

Sex Female Female Female Female

Age 75 years 39 years 39 years 47 years

Gross Findings Tumor size 8 times 8 cm ill demarcated Cut section firm

Tumor size 3 times 35 cm ill demarcated Cut section soft

and white

Tissue size 3 times 35 cm Tumor size 2 times 2 cm and 3 times 2 cm ill demarcated Cut section

firm and graywhite

Source Integrated Laboratory Services-Biotech (ILSbio) Chestertown MD 21620 httpwwwilsbiocom ILS-19106Detailed histopathology reports are available at the Protein BioTechnologies website httpwwwproteinbiotechnologiescom

Page 4 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

biomarkers of breast cancer For example Veeck and co-workers [9] showed recently that the extracellular matrixprotein Inter-α-trypsin inhibitor is strongly down-regu-lated in all human breast cancers

Thus in cancer proteomics especially where it is desiredto recover as much of the cellular or tissue proteins as pos-sible it is important not to completely rely on a singlelysis buffer that could exclude an entire class of criticallyneeded proteins especially the low abundance proteinsIndeed there are numerous proteomics publications thatfall exactly under the above category In breast cancer pro-teomics for example articles are found (example Refs[2325-30]) wherein researchers wanted to extract all pro-teins present in the breast tumor specimens but usedRIPA buffer as the sole buffer

In a study of ECM components in differentiating terato-carcinoma cells Grover and Adamson [31] noted thatfibronectin was poorly solubilized in RIPA buffer Other-wise no other direct reference suggesting insolubility ofECMs in RIPA buffer was found as yet Instead researchers(examples Refs [913-152032]) routinely use RIPAbuffer as the sole lysis and solubilization buffer in theimmunoprecipitation [15] immunoblot [131420] and

Western blot [932] analysis of extracellular matrix pro-teins

MethodsThe key steps (Figure 1) include protein extraction frombreast tumors and matched normal breast tissues sampleclean-up with GE Healthcare tools trypsin digestiondesalting with Michrom cartridges mass spectrometry2Dnano-LCESI-MSMS database search and protein IDdata processing and bioinformatics

Breast tumor tissue lysatesProtein lysates of breast tumors and matching normalbreast tissues are obtained from Protein BiotechnologiesInc (Ramona CA) The samples are selected to representdifferent types of tumors (T2-018 Tumor T2-048 TumorT2-029 Tumor) non-neoplastic normal breast tissue (T2-048 Normal) Diagnosis Stage Grade Age and GrossFindings (Table 1)

Protein extractionLysates provided by Protein Biotechnologies wereextracted by a two-step procedure First proteins areextracted with a modified Radio-ImmunoprecipitationAssay (RIPA) lysis buffer to yield the soluble fraction Sec-

Proteomics Workflow flow chartFigure 1Proteomics Workflow flow chartoutlining the key experimental steps in this work

Page 3 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

ond the residual insoluble fraction left after RIPA bufferextraction is subjected to additional extraction using aurea-based buffer to produce a second protein fractionThe compositions of the lysis buffers are as follows

Extraction 1Radioimmunoprecipitation Assay Lysis Buffer (modi-fied) Base Ingredients (PBS pH 74 SDS 01 Na-deox-ycholate 025) RIPA Protease Inhibitors (1 mMPhenylmethylsulfonyl fluoride (PMSF) EDTA 1 mM (cal-cium chelator 100 mM stock solution in H2O pH 74)Leupeptin 1 μgmL Aprotinin 1 μgmL Pepstatin 1 μgmL NaF (1 mM)) RIPA Phosphatase Inhibitors (Acti-vated Na3VO4 (1 mM))

Extraction 2Urea Lysis Buffer (PBS pH 74 50 M Urea 20 M Thiou-rea 50 mM DTT 01 SDS)

Cleanup of lysatesIn the cleanup step the proteins are separated from buff-ers detergents salts and other contaminants using amethod that is largely derived from a protocol and 2Dclean-up kit provided by Amersham Biosciences (GEHealthcare Piscataway NJ) The kit consists of four rea-gents a precipitant that precipitates the proteins to formpellets a co-precipitant that co-precipitates with the pro-teins and enhances their removal from solution a washbuffer that removes non-protein contaminants from theprotein precipitate and a wash additive that promotesrapid and complete re-suspension of the sample proteins

Prior to the beginning of clean-up the wash buffer waschilled at -20degC for 1 hr After thawing and spinning

down 100 μg aliquots of breast tumor lysates andmatched normal breast tissue lysates 300 μL of the pre-cipitant was added The mixture was vortexed on Eppen-dorf Thermomixer R (Eppendorf North AmericaWestbury NY) and then incubated in ice for 15 minutesNext 300 μL of co-precipitant was added and the mixturemixed The mixture was centrifuged at 12000 times g for 5minutes to pellet the proteins The clear supernatant liq-uid was carefully pipetted out while retaining the proteinprecipitate at the bottom of the 15 mL Eppendorf tubeWithout disturbing the pellet layer 40 μL of co-precipi-tant was added to the supernatant through the tilted sideof the 15 mL Eppendorf tube The mixture was kept in icefor 5 minutes before centrifuging it again at 12000 times g foranother 5 minutes The pellet was dispersed by adding 25μL of MilliQ water and centrifuging for 10 minutes Afteradding 1 mL of chilled wash buffer at -20degC and 5 μL ofwash additive the mixture was vortexed once every 30 sec-onds for a total of 35 minutes At this point the proteinsdid not dissolve but dispersed The mixture was againcentrifuged at 12000 times g for 5 minutes The supernatantwas carefully discarded and the pellet dried The pelletsare amorphous

In-solution digestionThe dried pellet was re-suspended in 20 μL 8 M urea100mM ammonium bicarbonate (ABC) and 06 μL of 100mM Dithiothreitol (DTT) in 100 mM ABC (ie 3 mMDTT) was stirred in Eppendorf Thermomixer R (Eppen-dorf North America Westbury NY) for 1 hr at 29degC Afteradjusting to room temperature 15 μL of 200 mM iodoa-cetamide (IAA) in 100 mM ABC (final concentration of 15mM IAA) was added Alkylation was then carried out byincubating the mixture for 45 minutes in a darkroom

Table 1 Characteristics of the breast tumors and matched normal breast tissues analyzed in this work

T2-018 Tumor T2-048 Tumor T2-048 Normal T2-029 Tumor

Location Right breast Left breast Left breast Bilateral

Diagnosis Infiltrating ductal carcinoma Infiltrating ductal carcinoma Normal Adenocarcinoma

Stage IV T4N0M0 IIB T2N1M0 Normal IV T4N1M0

Grade II I I III

Sex Female Female Female Female

Age 75 years 39 years 39 years 47 years

Gross Findings Tumor size 8 times 8 cm ill demarcated Cut section firm

Tumor size 3 times 35 cm ill demarcated Cut section soft

and white

Tissue size 3 times 35 cm Tumor size 2 times 2 cm and 3 times 2 cm ill demarcated Cut section

firm and graywhite

Source Integrated Laboratory Services-Biotech (ILSbio) Chestertown MD 21620 httpwwwilsbiocom ILS-19106Detailed histopathology reports are available at the Protein BioTechnologies website httpwwwproteinbiotechnologiescom

Page 4 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

ond the residual insoluble fraction left after RIPA bufferextraction is subjected to additional extraction using aurea-based buffer to produce a second protein fractionThe compositions of the lysis buffers are as follows

Extraction 1Radioimmunoprecipitation Assay Lysis Buffer (modi-fied) Base Ingredients (PBS pH 74 SDS 01 Na-deox-ycholate 025) RIPA Protease Inhibitors (1 mMPhenylmethylsulfonyl fluoride (PMSF) EDTA 1 mM (cal-cium chelator 100 mM stock solution in H2O pH 74)Leupeptin 1 μgmL Aprotinin 1 μgmL Pepstatin 1 μgmL NaF (1 mM)) RIPA Phosphatase Inhibitors (Acti-vated Na3VO4 (1 mM))

Extraction 2Urea Lysis Buffer (PBS pH 74 50 M Urea 20 M Thiou-rea 50 mM DTT 01 SDS)

Cleanup of lysatesIn the cleanup step the proteins are separated from buff-ers detergents salts and other contaminants using amethod that is largely derived from a protocol and 2Dclean-up kit provided by Amersham Biosciences (GEHealthcare Piscataway NJ) The kit consists of four rea-gents a precipitant that precipitates the proteins to formpellets a co-precipitant that co-precipitates with the pro-teins and enhances their removal from solution a washbuffer that removes non-protein contaminants from theprotein precipitate and a wash additive that promotesrapid and complete re-suspension of the sample proteins

Prior to the beginning of clean-up the wash buffer waschilled at -20degC for 1 hr After thawing and spinning

down 100 μg aliquots of breast tumor lysates andmatched normal breast tissue lysates 300 μL of the pre-cipitant was added The mixture was vortexed on Eppen-dorf Thermomixer R (Eppendorf North AmericaWestbury NY) and then incubated in ice for 15 minutesNext 300 μL of co-precipitant was added and the mixturemixed The mixture was centrifuged at 12000 times g for 5minutes to pellet the proteins The clear supernatant liq-uid was carefully pipetted out while retaining the proteinprecipitate at the bottom of the 15 mL Eppendorf tubeWithout disturbing the pellet layer 40 μL of co-precipi-tant was added to the supernatant through the tilted sideof the 15 mL Eppendorf tube The mixture was kept in icefor 5 minutes before centrifuging it again at 12000 times g foranother 5 minutes The pellet was dispersed by adding 25μL of MilliQ water and centrifuging for 10 minutes Afteradding 1 mL of chilled wash buffer at -20degC and 5 μL ofwash additive the mixture was vortexed once every 30 sec-onds for a total of 35 minutes At this point the proteinsdid not dissolve but dispersed The mixture was againcentrifuged at 12000 times g for 5 minutes The supernatantwas carefully discarded and the pellet dried The pelletsare amorphous

In-solution digestionThe dried pellet was re-suspended in 20 μL 8 M urea100mM ammonium bicarbonate (ABC) and 06 μL of 100mM Dithiothreitol (DTT) in 100 mM ABC (ie 3 mMDTT) was stirred in Eppendorf Thermomixer R (Eppen-dorf North America Westbury NY) for 1 hr at 29degC Afteradjusting to room temperature 15 μL of 200 mM iodoa-cetamide (IAA) in 100 mM ABC (final concentration of 15mM IAA) was added Alkylation was then carried out byincubating the mixture for 45 minutes in a darkroom

Table 1 Characteristics of the breast tumors and matched normal breast tissues analyzed in this work

T2-018 Tumor T2-048 Tumor T2-048 Normal T2-029 Tumor

Location Right breast Left breast Left breast Bilateral

Diagnosis Infiltrating ductal carcinoma Infiltrating ductal carcinoma Normal Adenocarcinoma

Stage IV T4N0M0 IIB T2N1M0 Normal IV T4N1M0

Grade II I I III

Sex Female Female Female Female

Age 75 years 39 years 39 years 47 years

Gross Findings Tumor size 8 times 8 cm ill demarcated Cut section firm

Tumor size 3 times 35 cm ill demarcated Cut section soft

and white

Tissue size 3 times 35 cm Tumor size 2 times 2 cm and 3 times 2 cm ill demarcated Cut section

firm and graywhite

Source Integrated Laboratory Services-Biotech (ILSbio) Chestertown MD 21620 httpwwwilsbiocom ILS-19106Detailed histopathology reports are available at the Protein BioTechnologies website httpwwwproteinbiotechnologiescom

Page 4 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Then 15 μL of 200 mM DTT100 mM ABC was added toconsume any un-reacted IAA The urea concentration wasreduced to about 1 M by diluting the mixture with 140 μLof (50 mM ABC + 2 mM CaCl2) Digestion was carried outby adding 6 μL of 040 μgμL = 24 μg of PromegaSequencing Grade trypsin and incubating in EppendorfThermomixer R for 20 hr at 374degC

At the end of the 20-hr incubation reaction was stoppedby adding 40 μL of 2 acetonitrile Then 6 μL of 10TFA was added to adjust the pH to 50

DesaltingManual Micro trap desalting cartridge and protocol fromMichrom (Michrom BioResources Auburn CA) are usedFirst the micro Trap is washed with 80 μL of LCMS Sol-vent B (90ACN01 TFA) Next it is equilibrated with80 μL of LCMS Solvent A (2 ACN01 TFA) Then 20μL of peptide digest sample is loaded onto the micro Trapsalts are removed by washing with 50 μL aliquots of LCMSsolvent A (2 ACN01 TFA) Tryptic peptides areeluted from the micro Trap with 16 μL of 70 ACNDesalted peptides are evaporated to dryness on an SC2SpeedVACreg Plus Thermo savant (Thermo Fisher ScientificWaltham MA)

Multidimensional nanoHPLCThe nano-HPLC is a Paradigm MS4B Multi-DimensionalHPLC equipped with a Michrom Paradigm AS1 refriger-ated autosampler and χCalibur software plugin (MichromBioResources Auburn CA) It is configured and operatedin a 3-1 column-switching arrangement All four pumpsare used pump D is used for sample loading onto captrapcartridge (sample concentration and de-salting) at 50 μLminute for 5 minutes Solvents A is 100 MilliQ waterno additives added Solvent B is HPLC Grade acetonitrile(ACN) from Burdick and Jackson (Honeywell Burdick ampJackson Morristown NJ) no additives added Solvent Cis 100 mL of FAHFBA mix + 900 mL water (FA = formicacid HFBA = HPLC Grade Ionatetrade Heptafluorobutyricacid from Pierce (Pierce Biotechnologies Rockford IL))The FAHFBA mix consists of 10 FA + 05 HFBA +895 water Solvent D is made by mixing 10 mL FAHFBA mix + 20 mL ACN + 970 mL water

The stock solution used to make up the MudPIT salt solu-tions is prepared as follows 5 ACN + 90 water + 25formic acid + 25 ammonium hydroxide + 005 HFBAThe salt plugs for MudPIT analysis are then prepared byserial dilution of this stock solution with solvent Ddescribed above

An optimized 60-minute nano gradient consists of the fol-lowing settings

Time (min) Flow rate (μlmin) B () C () (000 030500 1000 500 030 500 1000 1200 030 15001000 4700 030 4000 1000 5500 030 60001000 5600 030 8000 1000 5800 030 80001000 5900 030 500 1000 6500 030 5001000)

The five Solvent System and Event Group 2(D2) pulsesequences for autosampler mass spec and solvent flow asfollows Time (min) Event (000 Valve 1 inject 005Valve 2 inject 500 Valve 2 Load 510 Start Mass Spec 000 Valve 1 inject) whereas three Solvent System andEvent Group 1(D1) settings for sample concentration anddesalting on trap column are Time (min) flow (μLmin) (000 50 500 50 510 500)

Columns (Michrom BioResources)bull Peptide Nanotrap (TR52510942) 150 μm times 50 mm400 nL volume

bull Nanotrap analytical column (CL56124100) 5 μm 200Magic C18 75 μm times 150 mm

bull SCX Captrap (TR12510835) Contains a mediumpore large particle silica-based strong cation exchangematerial (PolySulfoethyl Aspartamide) Binds proteindigests peptides and other molecules (05ndash50 kD) for 1Dor 2D analysis Concentrates samples up to 100 fold (pHrange 27ndash70)

Nano-LCMSMSOne-dimensional (1D) and two-dimensional (2D) nanoHPLC experiments are run at a flow rate of 300 nLminSamples are loaded onto trap columns for concentrationand desalting at 50 μLmin For each experiment 12 μL ofpeptide digest resulting from 50ndash100 μg total protein isused 2 μL is injected for 1D 10 μL for 2D Each shotgunexperiment consists of a 12-cycle MudPIT run in which a60-minute nano-LC gradient is run for each of 1D 2D2D (0 mM NH4COO) 2D (25 mM NH4COO) 2D (50mM NH4COO) 2D (75 mM NH4COO) 2D (100 mMNH4COO) 2D (150 mM NH4COO) 2D (200 mMNH4COO) 2D (250 mM NH4COO) 2D (300 mMNH4COO) and 2D (500 mM NH4COO)

NanosprayThe nanospray is a Paradigm Nanotrap Platform(Michrom BioResources Auburn CA) equipped with aParadigm Metal spray needle The spray tip is a 75 cmlong 30 μm (Internal Diameter) times 105μm (Outer Diame-ter) surgical stainless steel electrochemically cut and pol-ished and sheathed by a 125 μm PEEK Tubing Theneedle permits flow range of 05 to 10 μLmin and a volt-age range of 1000 to 5000 Volts A 116 stainless steelValco nut attaches the spray needle to a 116 to 116

Page 5 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Valco union which is mounted on the Nanotrap Plat-form

Nanospray source parametersSheath Gas Flow Rate = 0 Aux Gas Flow Rate = 0 SprayVoltage (kC) = 251 Spray Current (μA) = -005 CapillaryTemp (degC) = 22110 Capillary Voltage (V) = 922 TubeLens Offset (V) = 50

Mass spectrometryData-dependent MS and MSMS spectra are acquired onan LCQ Deca Xp plus (Thermo Fisher Scientific San JoseCA)

MS and MSMSFive scan events are recorded for each data acquisitioncycle The first scan event is used for full scan MS acquisi-tion from 300ndash1800 mz Data are recorded in the cen-troid mode only (scan event 1 does not permit profilemode of data acquisition) The remaining four scan eventsare used for Collisionally Activated Decomposition(CAD) the four most abundant ions in each MS areselected and fragmented to produce product ion massspectra All CAD product ions are recorded in the profilemode

Ion optics settings usedMultipole 1 Offset (V) = -615 Lens voltage (V) = -2423Multipole 2 Offset (V) = -1054 Multipole RF Amplitude(Vp-p) = 400 Entrance Lens (V) = -5079

MSMS parametersNumber of microscans 4 Maximum Injection Time (ms)200 Isolation width (mz) 3 Normalized CollisionEnergy () 35 Activation Q 0250 Activation Time(msec) 3000 Scan Range 300ndash1800 mz

Database searches and protein identificationProteins are identified by searching the MS and MSMSspectra against NCBI nr human fasta using Bioworks v32(Thermo Fisher Scientific Inc San Jose CA) The UnifiedSearch Results File format (SRF) is employed rather thanthe traditional SEQUEST DTA and OUT formats Pep-tide and protein hits are scored and ranked using the newprobability-based scoring algorithm and the new finalScore (Sf) that are incorporated in Bioworks 32 BecauseMSMS spectra are acquired in the profile mode eachnano-LCMSMS run range in size from 95ndash120 Mega-bytes depending on the number of microscans (typically4 ms) in the Advanced Define Scan function in the MSMSLCQtune file of the Instrument Method

FiltersOnly peptides identified as possessing fully tryptic termini(containing up to two missed internal trypsin cleavage

sites) with cross-correlation scores (Xcorr) greater than 19for singly charged peptides 23 for doubly charged pep-tides and 375 for triply charged peptides are used forpeptide identification In addition the delta-correlationscores (ΔCn) must be greater than 01 for peptide identifi-cation Protein probability P(pro) le 0001

BioinformaticsBioinformatics calculations are carried out usingBlast2GO [33] Version 127 httpweb3vs160142vserverde a java webstart-enabled GeneOntology annotation visualization and analysis softwareThe calculations as implemented here consist of threekey sequential steps (a) Basic Local Alignment SearchTool (BLAST) [3435] (b) Mapping and (c) Annotation

BLASTIn BLAST [34] protein input queries are submitted to theBLAST server at the National Center for BiotechnologyInformation (NCBI) of the National Institutes of Health(NIH) over the internet httpwwwncbinlmnihgovblastBlastcgi The BLAST server generates hits (hit geneids (gi) and gene namesaccessions) The input parame-ters used are as follows BLAST database NCBI nr numberof BLAST hits requested for each query 40 BLAST expect-Value (ie eValue) 1e-3 BLAST program blastp Blast Ver-sions BLASTP 2213 [Nov-27-2005] and BLASTP 2215[Oct-15-2006] Blast Mode QBlast-NCBI HSP length cut-off 33 (please see The NCBI Handbook[36] for moreinformation on BLAST parameters)

The blast server accepts only fasta-formatted proteinsequences as input queries Although Bioworks 32 orlater can convert protein sequences into fasta text files theprotein sequences must be submitted from within Biow-orks browser prior to exiting the initial protein identifica-tion step Thus batch conversion of protein queries post-Bioworks is not possible via Bioworks route For exampleBioworks would not allow the analysis of only the pro-teins that occur in both tumor and normal because thesemust be determined post-Bioworks protein identificationAnother approach is Batch Entrez at the NCBI websitehttpwwwncbinlmnihgoventrezbatchentrezcgidb=Nucleotide) with Protein database selectedto import the batch file and displaying all in fasta format

For each of the protein input queries the BLAST machinegenerates a BLAST Table [34] exemplified here (Addi-tional File 1) with discoidin domain receptor 1 (DDR1)gi|68533097|dbj|BAE061031|894 (DDR1 is found inT2-081-T1 T2-018T2 T2-029-T2 T2-048N1 and T2-048N2 (Table 1)) The BLAST Table shows parameters ofthe BLAST search Sequences producing significant align-ments Gene Name Accession number e-Value align-

Page 6 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Valco union which is mounted on the Nanotrap Plat-form

Nanospray source parametersSheath Gas Flow Rate = 0 Aux Gas Flow Rate = 0 SprayVoltage (kC) = 251 Spray Current (μA) = -005 CapillaryTemp (degC) = 22110 Capillary Voltage (V) = 922 TubeLens Offset (V) = 50

Mass spectrometryData-dependent MS and MSMS spectra are acquired onan LCQ Deca Xp plus (Thermo Fisher Scientific San JoseCA)

MS and MSMSFive scan events are recorded for each data acquisitioncycle The first scan event is used for full scan MS acquisi-tion from 300ndash1800 mz Data are recorded in the cen-troid mode only (scan event 1 does not permit profilemode of data acquisition) The remaining four scan eventsare used for Collisionally Activated Decomposition(CAD) the four most abundant ions in each MS areselected and fragmented to produce product ion massspectra All CAD product ions are recorded in the profilemode

Ion optics settings usedMultipole 1 Offset (V) = -615 Lens voltage (V) = -2423Multipole 2 Offset (V) = -1054 Multipole RF Amplitude(Vp-p) = 400 Entrance Lens (V) = -5079

MSMS parametersNumber of microscans 4 Maximum Injection Time (ms)200 Isolation width (mz) 3 Normalized CollisionEnergy () 35 Activation Q 0250 Activation Time(msec) 3000 Scan Range 300ndash1800 mz

Database searches and protein identificationProteins are identified by searching the MS and MSMSspectra against NCBI nr human fasta using Bioworks v32(Thermo Fisher Scientific Inc San Jose CA) The UnifiedSearch Results File format (SRF) is employed rather thanthe traditional SEQUEST DTA and OUT formats Pep-tide and protein hits are scored and ranked using the newprobability-based scoring algorithm and the new finalScore (Sf) that are incorporated in Bioworks 32 BecauseMSMS spectra are acquired in the profile mode eachnano-LCMSMS run range in size from 95ndash120 Mega-bytes depending on the number of microscans (typically4 ms) in the Advanced Define Scan function in the MSMSLCQtune file of the Instrument Method

FiltersOnly peptides identified as possessing fully tryptic termini(containing up to two missed internal trypsin cleavage

sites) with cross-correlation scores (Xcorr) greater than 19for singly charged peptides 23 for doubly charged pep-tides and 375 for triply charged peptides are used forpeptide identification In addition the delta-correlationscores (ΔCn) must be greater than 01 for peptide identifi-cation Protein probability P(pro) le 0001

BioinformaticsBioinformatics calculations are carried out usingBlast2GO [33] Version 127 httpweb3vs160142vserverde a java webstart-enabled GeneOntology annotation visualization and analysis softwareThe calculations as implemented here consist of threekey sequential steps (a) Basic Local Alignment SearchTool (BLAST) [3435] (b) Mapping and (c) Annotation

BLASTIn BLAST [34] protein input queries are submitted to theBLAST server at the National Center for BiotechnologyInformation (NCBI) of the National Institutes of Health(NIH) over the internet httpwwwncbinlmnihgovblastBlastcgi The BLAST server generates hits (hit geneids (gi) and gene namesaccessions) The input parame-ters used are as follows BLAST database NCBI nr numberof BLAST hits requested for each query 40 BLAST expect-Value (ie eValue) 1e-3 BLAST program blastp Blast Ver-sions BLASTP 2213 [Nov-27-2005] and BLASTP 2215[Oct-15-2006] Blast Mode QBlast-NCBI HSP length cut-off 33 (please see The NCBI Handbook[36] for moreinformation on BLAST parameters)

The blast server accepts only fasta-formatted proteinsequences as input queries Although Bioworks 32 orlater can convert protein sequences into fasta text files theprotein sequences must be submitted from within Biow-orks browser prior to exiting the initial protein identifica-tion step Thus batch conversion of protein queries post-Bioworks is not possible via Bioworks route For exampleBioworks would not allow the analysis of only the pro-teins that occur in both tumor and normal because thesemust be determined post-Bioworks protein identificationAnother approach is Batch Entrez at the NCBI websitehttpwwwncbinlmnihgoventrezbatchentrezcgidb=Nucleotide) with Protein database selectedto import the batch file and displaying all in fasta format

For each of the protein input queries the BLAST machinegenerates a BLAST Table [34] exemplified here (Addi-tional File 1) with discoidin domain receptor 1 (DDR1)gi|68533097|dbj|BAE061031|894 (DDR1 is found inT2-081-T1 T2-018T2 T2-029-T2 T2-048N1 and T2-048N2 (Table 1)) The BLAST Table shows parameters ofthe BLAST search Sequences producing significant align-ments Gene Name Accession number e-Value align-

Page 6 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

length positives similarity hsp and mapping (Ontol-ogies found) for each of the 40 hits requested

MappingIn the Mapping step various databases are searched toidentify and fetch Gene Ontologies (GO) associated withthe hits obtained from NCBI BLAST searches

AnnotationThe annotation procedure selects the GO terms from theGO pool obtained by the mapping step and assigningthem to the query sequences using Annotation RuleAnnotation parameters are Pre-eValue-Hit-Filter 6 Pre-Similarity-Hit-Filter 30 Annotation Cut-Off 55 GO-Weight 5

Annotations are validated and expanded using an annota-tion expander The expander developed by a group at theNorwegian University of Science and Technology httpwwwgoatno deploys an additional Gene Ontologystructure the Second Gene Ontology Layer to suggestnew Biological Processes and Cellular Componentsbased on the genes existing Molecular Function annota-tions

ResultsThe multidimensional protein identification technology(MudPIT) mass spectra of the breast specimens T2-018Tumor T2-048 Tumor T2-048 Normal and T2-029Tumor are shown in Additional Files 2 3 4 and 5 respec-

tively The set of 12-cycle spectra to the left of the figuresare the RIPA buffer fractions whereas the spectra of the 12urea buffer fractions are shown at right

The mass spectra clearly show that for each of the 60-minute MudPIT runs the 1D_2 μL and 2D_10 μL spectraappear to produce higher ion currents than the rest of thespectra This may be partly due to the greater concentra-tions of peptides in these runs in the 1D_2 μL run all two2 μL of approximately 5 μg total peptide digest are intro-duced into the nano-LCESI-MSMS system via the pep-tide Nanotrap and analytical column (the SCX column isbypassed for 1D analysis) And in 2D_10 μL all 10 μL ofpeptide digest are deposited on the SCX column allunbound peptides are washed into the mass spectrometerafter pre-concentration and de-salting at the 150 μm times 50mm peptide (40 nanoliter volume) nanotrap Thus theamounts of sample introduced into the mass spectrome-ter in these two runs maybe responsible for the higher ioncurrents

The MudPIT spectra also show that the RIPA-insolublematerials contain quite a significant amount of proteinsas reflected in the highly abundant spectra of urea-solubleproteins This clearly raises concern about using RIPAbuffer as the sole lysis buffer for the proteomics of breastcancer (or other cancers as well)

The proteins identified in the database search of the Mud-PIT mass spectra are shown in Table 2 Here T2-048T

Table 2 Proteins found in the MudPIT proteomics of the RIPA and urea fractions of the breast tumors analyzed in this work

Proteins Proteins_validated

Lysate Default bull Xcorr (123) 190 23 375bull P(pro) 0001 bull ΔCn ge 01

Specimen RIPA Buffer Urea Buffer

T2-018_Tumor 33147 299

T2-018_Tumor 31828 155

T2-048_Tumor 27049 143

T2-048_Tumor 31329 261

T2-048_Tumor 28203 195

T2-048_Tumor 30816 122

T2-029_Tumor 27944 215

T2-029_Tumor 24963 358

Page 7 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

(RIPA) and T2-048N2 (UREA) represent the RIPA-solu-ble fraction of the tumor sample T2-048 and the urea-sol-uble fraction of the normal breast tissue sample T2-048respectively The column labelled Default depicts thedefault number of proteins found by Bioworks 32 priorto application of filter functions and protein validationThe final sets of proteins identified are shown at the right-most column Again it is clear that the RIPA-insolublematerials which dissolve in urea contain significantamounts of proteins that would otherwise be discarded ifRIPA buffer was the only lysis and solubilization bufferused

The partitioning of the proteins between RIPA and ureabuffers in the infiltrating Ductal Carcinoma case T2-018TUMOR (Figure 2A) shows that the average molecularweight of the 217 proteins that dissolve exclusively inRIPA buffer was 61604 mz whereas the 73 proteins thatdissolve exclusively in urea buffer have an average molec-ular weight of 99154 mz That is the average molecularweight of proteins that dissolve exclusively in urea was37551 mz or 61 higher than in RIPA buffer Finally theaverage molecular weight of the 82 proteins that dissolvein both RIPA and urea buffers was 40490 mz

The corresponding data for T2-048 TUMOR T2-048NORMAL and T2-029 TUMOR are shown in Venn Dia-grams in Figures 2B C and 2D

Gene OntologyIn an effort to determine the protein groups and identifythe fractionated proteins contained in the compartmentsshown in Table 2 and Figure 2 Gene Ontology[3738]httpwwwgeneontologyorg was employed

The Gene Ontology (GO) project which began in 1998 asa collaboration between three databases FlyBase Saccha-romyces Genome Database and the Mouse GenomeDatabase has today grown to encompass nearly all majordatabases and has become a new powerful tool for min-ing biology data Gene Ontology annotates biologicaldata in terms of their Biological Process MolecularFunction and Cellular Component

Biological Process is an ensemble of biochemical trans-formations that are accomplished by one or more orderedassemblies of molecular functions Biological Process maybe broad (physiological process metabolism etc) or spe-cific (nitric oxide metabolism oxygen transport etc)

Molecular Function is the specific elemental action ortask performed by a gene product or assembled complexesof gene products Examples of broad molecular functionsinclude catalysis binding and structural molecular activ-ity whereas specific molecular functions are exemplified

by tetrapyrrole binding adenylate cyclase activity and cal-modulin binding

Cellular Component is the subcellular location(organelle nucleus etc) and macromolecular complexeswere the gene product is located

In this work proteins identified by proteomic analyses aresubmitted to GO analyses including NCBI BLAST map-ping and annotation The results are presented as aDirected Acyclic Graph (DAG) which shows the numberof annotated sequences and the annotation scores con-tributing to each node The nodes are color-coded andthe relative importance of each annotation score is indi-cated by the intensity of the orange color at that nodeThere are three types of nodes a double-edged octagonrepresents an annotated GO term a rectangle represents anon-annotated GO term node and an oval shape denotesGene Ontology obtained by mapping which can directlybe directly associated to one or more BLAST hits

In the Biological Process formulation functionalities thatdirectly identify the proteins in the RIPA and urea bufferfractions are not explicitly evident The Molecular Func-tion annotations and especially the Cellular Compo-nent however do provide highly useful information onprotein groups and identities of the protein fractionsshown in the Venn diagrams of Figure 2 In the followingsections the Cellular Component annotations will be pri-marily used to characterize the protein fractionsalthough the Molecular Functions annotations will alsobe used

Cellular ComponentT2-018T (RIPA) and T2-018T (Urea) fractionsThe Cellular Component DAGs for specimens T2-018T(RIPA) and T2-018T (Urea) are shown in Figures 3 and 4respectively Interestingly the Cellular Component parentnode of T2-018T DAG shows that mapping found curatedGene Ontologies (Ontologies) for 262 out of the original299 proteins found in this sample (Table 2) Similarly inT2-018T 131 out of the 155 proteins have Ontologiesavailable in the Go databases The high percentages ofproteins that have curated Ontologies thus provide ade-quate bioinformatics data needed to characterize the RIPAand urea proteomes with high degree of specificity andreliability

Comparison of the T2-018T (RIPA) and T2-018T (Urea)DAGs shows that the entire set of extracellular matrix pro-tein nodes in the urea (T2-018T Figure 4) DAG are almostcompletely missing from the RIPA (T2-018T Figure 3)DAG at the indicated node filter settings suggesting thatnearly all extracellular matrix proteins are dissolved in theurea but not in the RIPA buffer

Page 8 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 9 of 24(page number not for citation purposes)

(A-D) A The partitioning of the proteins between RIPA and urea buffers in the infiltrating Ductal Carcinoma case T2-018T (TUMOR)Figure 2(A-D) A The partitioning of the proteins between RIPA and urea buffers in the infiltrating Ductal Carcinoma case T2-018T (TUMOR) Venn diagram showing that RIPA and urea lysis buffers fractionate breast tumor proteins primarily on the basis of molecular weights The average molecular weight of the 217 proteins that dissolve exclusively in RIPA buffer is 61604 mz whereas the 73 proteins that dissolve exclusively in urea buffer have an average molecular weight of 99154 mz B Differential partitioning of proteins extracted from the T2-048 TUMOR between RIPA and urea buffers Venn diagram showing that RIPA and urea lysis buffers fractionate breast tumor proteins on the basis of molecular weights The average molecular weight of the 69 proteins that dissolve almost exclusively in RIPA buffer is 72330 mz whereas the 187 proteins that dissolve exclusively in urea buffer have an average molecular weight of 79704 mz Seventy-four proteins whose average molecular weight is 47854mz dissolve fully in both RIPA and urea buffers C Matched normal breast tissue T2-048 NORMAL between RIPA and UREA buffers The average molecular weight of the 128 proteins that are soluble nearly exclusively in RIPA buffer is 61879 mz whereas the 55 proteins that dissolve exclusively in urea buffer have an average molecular weight of 93810 mz Sixty-seven proteins with an average molecular weight of 53102mz are fully soluble in both RIPA and urea buffers D Bilateral Adenocarcinoma case T2-029 TUMOR The average molecular weight of the 104 proteins that dissolve almost exclusively in RIPA buffer is 51485 mz whereas the 247 proteins that dissolve exclusively in urea buffer have an average molecular weight of 76092 mz One hundred and eleven proteins with an average molecular weight of 45126 mz are fully soluble in both RIPA and urea buffers

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

The node filter is a mechanism of simplifying otherwisecomplicated DAGs At a node filter of 28 for example allnodes whose number of annotated protein sequences are28 or below are not displayed n an effort to simplify theDAG Thus when the node filter is lowered previouslyhidden nodes are displayed making the chart morecrowded

Close-up sections of the extracellular regions (Figure 5)clearly show that extracellular matrix proteins dissolvealmost exclusively in urea buffer The cropped sections areobtained with node filters of 15 and 10 for the RIPA andurea DAGs respectively

Interestingly lowering the DAG node filter for the RIPADAG did not produce appreciable change in the numberof nodes displayed within the extracellular region shownin Figure 3 whereas even a slight lowering of the node fil-ter in the urea DAG (Figure 4) reveals a large number ofpreviously hidden nodes shown here in Figure 6 whenthe urea node filter is reduced to zero Again this is a fur-ther confirmation that most of the extracellular matrixproteins are dissolved in the urea buffer

Comparison also shows that for extracellular regionRIPA buffer has a greater number of annotated proteinsequences and annotation score than Urea buffer [RIPA

(extracellular region Seqs62 Score4087) UREA (extra-cellular region Seqs45 Score3631)] However thisgreater enrichment at the RIPA buffer node may indeed bedue to the greater number of proteins at the RIPA bufferCellular Component parent node [RIPA buffer (CellularComponent Seqs262 Score1185) UREA buffer (Cel-lular Component Seqs131 Score6259)] To correct forthis and allow for sequence-independent comparison ofthe nodes regardless of the number of protein sequencesthat was submitted to GO analyses (Table 2) thesequences and scores are expressed as percentages of thesequences and scores present at the Cellular Componentparent node This process is known as normalizationThus comparison of the extracellular regions become[RIPA (extracellular region (Seqs237 Score345)) UREA (extracellular region (Seqs344 Score580))] Itcan now be stated that based on Seqs the extracellularregion is preferentially enriched in Urea buffer by 344when compared to RIPA buffer (237) Similar calcula-tions are shown below for Seqs of (RIPAUREA) respec-tively nuclear (298206) membrane (218214)intracellular (824702) protein complex (424344)organelle (660542) cytoskeleton (187252) and cyto-plasmic proteins (607481) From here it is clear thatcytoskeletal and extracellular region proteins are selec-tively enriched in urea buffer whereas nuclear intracellu-lar protein complexes organelle cytoplasmic and

The Cellular Component DAG for the proteome T2-018T (RIPA)Figure 3The Cellular Component DAG for the proteome T2-018T (RIPA) Extracellular matrix proteins are not observed even at a node filter setting of 28

Page 10 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

mitochondrial proteins are preferentially enriched in theRIPA fractions

Selective protein enrichment comparisons based on bothSeqs and Scores are shown in Figures 7 and 8

T2-048T (RIPA) and T2-048T (Urea) fractionsRather than show the entire DAGs cropped views of theextracellular regions of T2-048T (RIPA) and T2-048T(Urea) DAGs (Figure 9) show that extracellular matrixproteins are present almost exclusively in the urea frac-tion It is also seen that mapping found that nearly 90of the original proteins present in the T2-048T proteome(ie 232 of 261 Table 2) have existing Ontologies thusproviding the requisite bioinformatics informationneeded for the characterization of the proteins SimilarlyOntologies are found for 120 of the 143 proteins (84)of the T2-048T proteome

T2-048N1 (RIPA) and T2-048N2 (Urea) fractionsHere again the extracellular regions of T2-048N1 (RIPA)and T2-048N2 (Urea) DAGs (Figure 10) show that extra-cellular matrix proteins are present nearly exclusively inurea buffer with 86 and 80 of the proteins of T2-

048N1 and T2-048N2 proteomes respectively beingfound to have existing Ontologies available It is interest-ing to compare the normalized extracellular matrix Seqsand scores of T2-048T (Tumor) and T2-048N2 (Normal)[T2-048T-UREA (extracellular matrix (Seqs1336Score1680)) T2-048N2-UREA (extracellular matrix(Seqs2959 Score3266))] The normalized scores andSeqs belonging to normal breast tissues are more than twotimes those of the breast tumors which may be due to thedegradation of extracellular matrix proteins in the tumorby proteases to pave the way for the metastasis of cancercells [8-22]

T2-029T1 (RIPA) and T2-029T2 (Urea) fractionsThe cropped-out extracellular regions displayed side-by-side (Figure 11) again show that extracellular matrix pro-teins are found nearly exclusively in urea buffer Andmapping found Ontologies for 84 and 87 of the T2-029T1 and T2-029T2 proteins respectively Upon lower-ing the node filter on the T2-029T2 DAG to zero a fulldisplay of the extracellular matrix proteins are obtained(Additional File 6) Similar lowering of the node filter onthe T2-029T1 DAG did not reveal significantly new or rel-evant structural information

The Cellular Component DAG for the proteome T2-018T (UREA)Figure 4The Cellular Component DAG for the proteome T2-018T (UREA) At a node filter setting of only 13 extracellular matrix proteins are highly evident Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 11 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Complete Cellular Component DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 7 8 9 10 11 and 12 respec-tively

Molecular FunctionMolecular Function annotations also contain elementsthat reveal protein solubilization preferences of the RIPAand urea buffers Thus far Structural Molecule Activity(SMA) is the only functionality in the Molecular Functionannotation that contains nodes that directly relate to theprotein solubility preferences In general the SMA node ofthe urea proteome contains a child node ExtracellularMatrix Structural Constituent which contains thenumber of annotated protein sequences and an annota-tion score for extracellular matrix proteins The SMAnode of the RIPA proteome does not contain any descrip-tors (at the given node filter settings) that suggest the pres-ence of Extracellular Matrix Structural ConstituentThus extracellular matrix proteins are observed nearlyexclusively in the urea fraction

The Molecular Function DAGs for T2-018T (RIPA) andT2-018T (UREA) (Figures 12 and 13 respectively) showclearly that Extracellular Matrix Structural Constituentsare present only in the urea proteome The RIPA and urea

Close-up views of the extracellular regions of T2-018T (UREA)Figure 5Close-up views of the extracellular regions of T2-018T (UREA) clearly show that extracellular matrix proteins dis-solve almost exclusively in urea buffer

Expanded view of the extracellular region of the cellular component DAG for the proteome T2-018T (UREA)Figure 6Expanded view of the extracellular region of the cel-lular component DAG for the proteome T2-018T (UREA) The node filter was reduced to 0 to obtain this complete display Lowering the DAG node filter for the RIPA DAG did not produce appreciable change in the number of nodes displayed within the extracellular region

Page 12 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

DAGs are drawn with node filter settings of 30 and 12respectively

In Figure 14(AndashB) are the cropped-out SMA nodes for T2-018T (RIPA) and T2-018T (UREA) when the node filtersare set to 2 and 1 respectively Further lowering of thenode filters did not produce significant differences in bothDAGs Also shown in Figure 14(CndashD) are the cropped-out SMA nodes for T2-029T1 (RIPA) and T2-029T2(UREA) proteomes Again extracellular matrix proteinsare found nearly exclusively in the urea fractions

The above trend is consistently maintained in Figure 15(A-D) for T2-048T (RIPA) and T2-048T (UREA) and T2-048N1 (RIPA) and T2-048N2 (UREA) respectively

Detailed Molecular Function DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 13 14 15 16 17 and 18respectively

DiscussionThe solubility of proteins in RIPA and urea buffersdepends on several physicochemical factors including the

characteristics of the proteins and properties of the RIPAand urea buffers

Physicochemical properties of the proteins that affecttheir solubility include average charge determined by therelative numbers of Asp Glu Lys and Arg residues andthe content of turn-forming residues (Asn Gly Pro andSer) [39] Insoluble proteins tend to have more hydro-phobic stretches longer than 20 amino acids residueslower glutamine content fewer negatively charged resi-dues and higher percentages of aromatic amino acid resi-dues than soluble ones [40] Indeed high contents ofnegatively charged amino-acid residues and absence ofhydrophobic patches tend to improve protein solubility[41] Also low percentage of aspartic acid glutamic acidasparagines and glutamine residues increases the proba-bility of a protein to be insoluble [41]

Solubility of proteins in lysis buffer also depends highlyon the composition and gross physicochemical propertiesof the lysis buffer These include [4243] the type ofbuffer the presence or absence of phosphate pH saltsampholytes detergents chaotropic agents reducingagents (dithiothreitol (DTT) dithioerythreitol (DTE) β-mercaptoethanol tributyl phosphine (TBP) tris-carboxy-lethylphosphate (TCEP))

Extraction of proteins from breast tumors for proteomic analysisFigure 7Extraction of proteins from breast tumors for pro-teomic analysis nuclear proteins (A) intracellular proteins (C) and protein complexes (D) are more soluble in RIPA buffer than in urea buffer RIPA whereas membrane proteins (B) are slightly more soluble in urea buffer

Extraction of proteins from breast tumors for proteomicsFigure 8Extraction of proteins from breast tumors for pro-teomics proteins of the extracellular region (A) and cytoskeleton (B) are more soluble in urea buffer than in RIPA whereas for cytoplasmic (C) and mitochondrial (D) proteins RIPA buffer is preferred

Page 13 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RIPA and UREA buffers fractionate breast cancer proteins primarily on the basis of molecular weightsRIPA buffer is a versatile and efficient lysis buffer suitablefor the recovery of most proteins including whole cellnuclear mitochondrial membrane receptors cytoskele-tal-associated and soluble proteins However as the datain Figures 2 shows there is a high molecular weight cut-off (ge 12 higher average molecular weight in urea thanRIPA) for a proteins solubility in RIPA buffer Proteins

with molecular weights of around 100 kDa or higher maynot dissolve readily unless they possess unique structuralfeatures that enhance their solubility in RIPA buffer Thuswhen a given protein group is subjected to RIPA bufferextraction the high molecular weight fraction may notdissolve ndash they are recovered as the RIPA-insoluble frac-tion that ultimately dissolve in urea buffer This mayexplain why such a high percentage of proteins are recov-ered in the urea fraction after they have resisted solubility

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA)Figure 9Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) Extracellular matrix proteins and virions are observed almost entirely in urea fraction

Page 14 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

in RIPA buffer (Figures 2D especially Figures 2D (T2-029(TUMOR))

Interestingly Ignatoski and co-workers [42] have alsodemonstrated that different lysis buffers solubilized dif-ferent subsets of cellular proteins (rather than entire pro-teins) based primarily on the molecular weights NeitherRIPA nor urea buffer was however used in this kinaseassay ndash they denature kinases All buffers that they testedwere non-denaturing

Nearly all extracellular matrix proteins are insoluble in RIPA buffer but dissolve readily in urea bufferPerhaps the most important finding in this work is thatnearly all extracellular matrix proteins (ECMs) are insolu-ble in RIPA buffer whereas they dissolve readily in ureabuffer This may be due to ECMs having very high molec-ular weights Why then is RIPA buffer being used rou-tinely to dissolve extracellular matrix proteins byresearchers especially in cancer research Indeed RIPAbuffer does dissolve high molecular weight proteins butthe recovery may be poor The solubility of a protein is acombination of many factors beyond the nature of thelysis buffer If for example the high molecular weightprotein has structural features that enhance its solubility(high contents of negatively charged amino-acid residuesand absence of hydrophobic patches [41]) as discussed inthe introduction section the protein would dissolve inRIPA buffer On the other a smaller molecular weight

protein may surprisingly fail to dissolve in RIPA buffer ifit aggregates or possesses structural features that hamperits solubility Some epigenetic post-translation or spon-taneous structural changes can also impede a proteinssolubility in RIPA buffer One example is tau whichwould normally be soluble in RIPA But when it becomeshyperphosphorylated for example by endogenouslyoverproduced Aβ protein in Alzheimers disease it wouldresist solubility in RIPA buffer [44]

Selective Enrichments of Protein Groups by RIPA and Urea BuffersData in Figure 7A shows that nuclear proteins are some-what more selectively enriched in RIPA buffer than inurea consistent with many standard molecular biologylaboratory practices RIPA buffer is one of the recom-mended (or one of the preferred) buffers for efficientrecovery of nuclear proteins [64546]

Protein complexes (Figure 7D) are also slightly more con-centrated in RIPA buffer than in urea Protein complexestend to have high molecular weights and although RIPAbuffer has poor solubility for high molecular weight pro-teins protein complexes are held together largely by non-covalent bonds Detailed Cellular Component DAGs(DAGs not shown) indicate that protein complexesreferred to here include immunoglobulin complexhemoglobin complex fibrinogen complex transcriptorfactor complex DNA polymerase complex DNA-directed

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA)Figure 10Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix proteins and virions are observed almost exclusively in urea fraction

Page 15 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

The node filter is a mechanism of simplifying otherwisecomplicated DAGs At a node filter of 28 for example allnodes whose number of annotated protein sequences are28 or below are not displayed n an effort to simplify theDAG Thus when the node filter is lowered previouslyhidden nodes are displayed making the chart morecrowded

Close-up sections of the extracellular regions (Figure 5)clearly show that extracellular matrix proteins dissolvealmost exclusively in urea buffer The cropped sections areobtained with node filters of 15 and 10 for the RIPA andurea DAGs respectively

Interestingly lowering the DAG node filter for the RIPADAG did not produce appreciable change in the numberof nodes displayed within the extracellular region shownin Figure 3 whereas even a slight lowering of the node fil-ter in the urea DAG (Figure 4) reveals a large number ofpreviously hidden nodes shown here in Figure 6 whenthe urea node filter is reduced to zero Again this is a fur-ther confirmation that most of the extracellular matrixproteins are dissolved in the urea buffer

Comparison also shows that for extracellular regionRIPA buffer has a greater number of annotated proteinsequences and annotation score than Urea buffer [RIPA

(extracellular region Seqs62 Score4087) UREA (extra-cellular region Seqs45 Score3631)] However thisgreater enrichment at the RIPA buffer node may indeed bedue to the greater number of proteins at the RIPA bufferCellular Component parent node [RIPA buffer (CellularComponent Seqs262 Score1185) UREA buffer (Cel-lular Component Seqs131 Score6259)] To correct forthis and allow for sequence-independent comparison ofthe nodes regardless of the number of protein sequencesthat was submitted to GO analyses (Table 2) thesequences and scores are expressed as percentages of thesequences and scores present at the Cellular Componentparent node This process is known as normalizationThus comparison of the extracellular regions become[RIPA (extracellular region (Seqs237 Score345)) UREA (extracellular region (Seqs344 Score580))] Itcan now be stated that based on Seqs the extracellularregion is preferentially enriched in Urea buffer by 344when compared to RIPA buffer (237) Similar calcula-tions are shown below for Seqs of (RIPAUREA) respec-tively nuclear (298206) membrane (218214)intracellular (824702) protein complex (424344)organelle (660542) cytoskeleton (187252) and cyto-plasmic proteins (607481) From here it is clear thatcytoskeletal and extracellular region proteins are selec-tively enriched in urea buffer whereas nuclear intracellu-lar protein complexes organelle cytoplasmic and

The Cellular Component DAG for the proteome T2-018T (RIPA)Figure 3The Cellular Component DAG for the proteome T2-018T (RIPA) Extracellular matrix proteins are not observed even at a node filter setting of 28

Page 10 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

mitochondrial proteins are preferentially enriched in theRIPA fractions

Selective protein enrichment comparisons based on bothSeqs and Scores are shown in Figures 7 and 8

T2-048T (RIPA) and T2-048T (Urea) fractionsRather than show the entire DAGs cropped views of theextracellular regions of T2-048T (RIPA) and T2-048T(Urea) DAGs (Figure 9) show that extracellular matrixproteins are present almost exclusively in the urea frac-tion It is also seen that mapping found that nearly 90of the original proteins present in the T2-048T proteome(ie 232 of 261 Table 2) have existing Ontologies thusproviding the requisite bioinformatics informationneeded for the characterization of the proteins SimilarlyOntologies are found for 120 of the 143 proteins (84)of the T2-048T proteome

T2-048N1 (RIPA) and T2-048N2 (Urea) fractionsHere again the extracellular regions of T2-048N1 (RIPA)and T2-048N2 (Urea) DAGs (Figure 10) show that extra-cellular matrix proteins are present nearly exclusively inurea buffer with 86 and 80 of the proteins of T2-

048N1 and T2-048N2 proteomes respectively beingfound to have existing Ontologies available It is interest-ing to compare the normalized extracellular matrix Seqsand scores of T2-048T (Tumor) and T2-048N2 (Normal)[T2-048T-UREA (extracellular matrix (Seqs1336Score1680)) T2-048N2-UREA (extracellular matrix(Seqs2959 Score3266))] The normalized scores andSeqs belonging to normal breast tissues are more than twotimes those of the breast tumors which may be due to thedegradation of extracellular matrix proteins in the tumorby proteases to pave the way for the metastasis of cancercells [8-22]

T2-029T1 (RIPA) and T2-029T2 (Urea) fractionsThe cropped-out extracellular regions displayed side-by-side (Figure 11) again show that extracellular matrix pro-teins are found nearly exclusively in urea buffer Andmapping found Ontologies for 84 and 87 of the T2-029T1 and T2-029T2 proteins respectively Upon lower-ing the node filter on the T2-029T2 DAG to zero a fulldisplay of the extracellular matrix proteins are obtained(Additional File 6) Similar lowering of the node filter onthe T2-029T1 DAG did not reveal significantly new or rel-evant structural information

The Cellular Component DAG for the proteome T2-018T (UREA)Figure 4The Cellular Component DAG for the proteome T2-018T (UREA) At a node filter setting of only 13 extracellular matrix proteins are highly evident Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 11 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Complete Cellular Component DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 7 8 9 10 11 and 12 respec-tively

Molecular FunctionMolecular Function annotations also contain elementsthat reveal protein solubilization preferences of the RIPAand urea buffers Thus far Structural Molecule Activity(SMA) is the only functionality in the Molecular Functionannotation that contains nodes that directly relate to theprotein solubility preferences In general the SMA node ofthe urea proteome contains a child node ExtracellularMatrix Structural Constituent which contains thenumber of annotated protein sequences and an annota-tion score for extracellular matrix proteins The SMAnode of the RIPA proteome does not contain any descrip-tors (at the given node filter settings) that suggest the pres-ence of Extracellular Matrix Structural ConstituentThus extracellular matrix proteins are observed nearlyexclusively in the urea fraction

The Molecular Function DAGs for T2-018T (RIPA) andT2-018T (UREA) (Figures 12 and 13 respectively) showclearly that Extracellular Matrix Structural Constituentsare present only in the urea proteome The RIPA and urea

Close-up views of the extracellular regions of T2-018T (UREA)Figure 5Close-up views of the extracellular regions of T2-018T (UREA) clearly show that extracellular matrix proteins dis-solve almost exclusively in urea buffer

Expanded view of the extracellular region of the cellular component DAG for the proteome T2-018T (UREA)Figure 6Expanded view of the extracellular region of the cel-lular component DAG for the proteome T2-018T (UREA) The node filter was reduced to 0 to obtain this complete display Lowering the DAG node filter for the RIPA DAG did not produce appreciable change in the number of nodes displayed within the extracellular region

Page 12 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

DAGs are drawn with node filter settings of 30 and 12respectively

In Figure 14(AndashB) are the cropped-out SMA nodes for T2-018T (RIPA) and T2-018T (UREA) when the node filtersare set to 2 and 1 respectively Further lowering of thenode filters did not produce significant differences in bothDAGs Also shown in Figure 14(CndashD) are the cropped-out SMA nodes for T2-029T1 (RIPA) and T2-029T2(UREA) proteomes Again extracellular matrix proteinsare found nearly exclusively in the urea fractions

The above trend is consistently maintained in Figure 15(A-D) for T2-048T (RIPA) and T2-048T (UREA) and T2-048N1 (RIPA) and T2-048N2 (UREA) respectively

Detailed Molecular Function DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 13 14 15 16 17 and 18respectively

DiscussionThe solubility of proteins in RIPA and urea buffersdepends on several physicochemical factors including the

characteristics of the proteins and properties of the RIPAand urea buffers

Physicochemical properties of the proteins that affecttheir solubility include average charge determined by therelative numbers of Asp Glu Lys and Arg residues andthe content of turn-forming residues (Asn Gly Pro andSer) [39] Insoluble proteins tend to have more hydro-phobic stretches longer than 20 amino acids residueslower glutamine content fewer negatively charged resi-dues and higher percentages of aromatic amino acid resi-dues than soluble ones [40] Indeed high contents ofnegatively charged amino-acid residues and absence ofhydrophobic patches tend to improve protein solubility[41] Also low percentage of aspartic acid glutamic acidasparagines and glutamine residues increases the proba-bility of a protein to be insoluble [41]

Solubility of proteins in lysis buffer also depends highlyon the composition and gross physicochemical propertiesof the lysis buffer These include [4243] the type ofbuffer the presence or absence of phosphate pH saltsampholytes detergents chaotropic agents reducingagents (dithiothreitol (DTT) dithioerythreitol (DTE) β-mercaptoethanol tributyl phosphine (TBP) tris-carboxy-lethylphosphate (TCEP))

Extraction of proteins from breast tumors for proteomic analysisFigure 7Extraction of proteins from breast tumors for pro-teomic analysis nuclear proteins (A) intracellular proteins (C) and protein complexes (D) are more soluble in RIPA buffer than in urea buffer RIPA whereas membrane proteins (B) are slightly more soluble in urea buffer

Extraction of proteins from breast tumors for proteomicsFigure 8Extraction of proteins from breast tumors for pro-teomics proteins of the extracellular region (A) and cytoskeleton (B) are more soluble in urea buffer than in RIPA whereas for cytoplasmic (C) and mitochondrial (D) proteins RIPA buffer is preferred

Page 13 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RIPA and UREA buffers fractionate breast cancer proteins primarily on the basis of molecular weightsRIPA buffer is a versatile and efficient lysis buffer suitablefor the recovery of most proteins including whole cellnuclear mitochondrial membrane receptors cytoskele-tal-associated and soluble proteins However as the datain Figures 2 shows there is a high molecular weight cut-off (ge 12 higher average molecular weight in urea thanRIPA) for a proteins solubility in RIPA buffer Proteins

with molecular weights of around 100 kDa or higher maynot dissolve readily unless they possess unique structuralfeatures that enhance their solubility in RIPA buffer Thuswhen a given protein group is subjected to RIPA bufferextraction the high molecular weight fraction may notdissolve ndash they are recovered as the RIPA-insoluble frac-tion that ultimately dissolve in urea buffer This mayexplain why such a high percentage of proteins are recov-ered in the urea fraction after they have resisted solubility

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA)Figure 9Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) Extracellular matrix proteins and virions are observed almost entirely in urea fraction

Page 14 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

in RIPA buffer (Figures 2D especially Figures 2D (T2-029(TUMOR))

Interestingly Ignatoski and co-workers [42] have alsodemonstrated that different lysis buffers solubilized dif-ferent subsets of cellular proteins (rather than entire pro-teins) based primarily on the molecular weights NeitherRIPA nor urea buffer was however used in this kinaseassay ndash they denature kinases All buffers that they testedwere non-denaturing

Nearly all extracellular matrix proteins are insoluble in RIPA buffer but dissolve readily in urea bufferPerhaps the most important finding in this work is thatnearly all extracellular matrix proteins (ECMs) are insolu-ble in RIPA buffer whereas they dissolve readily in ureabuffer This may be due to ECMs having very high molec-ular weights Why then is RIPA buffer being used rou-tinely to dissolve extracellular matrix proteins byresearchers especially in cancer research Indeed RIPAbuffer does dissolve high molecular weight proteins butthe recovery may be poor The solubility of a protein is acombination of many factors beyond the nature of thelysis buffer If for example the high molecular weightprotein has structural features that enhance its solubility(high contents of negatively charged amino-acid residuesand absence of hydrophobic patches [41]) as discussed inthe introduction section the protein would dissolve inRIPA buffer On the other a smaller molecular weight

protein may surprisingly fail to dissolve in RIPA buffer ifit aggregates or possesses structural features that hamperits solubility Some epigenetic post-translation or spon-taneous structural changes can also impede a proteinssolubility in RIPA buffer One example is tau whichwould normally be soluble in RIPA But when it becomeshyperphosphorylated for example by endogenouslyoverproduced Aβ protein in Alzheimers disease it wouldresist solubility in RIPA buffer [44]

Selective Enrichments of Protein Groups by RIPA and Urea BuffersData in Figure 7A shows that nuclear proteins are some-what more selectively enriched in RIPA buffer than inurea consistent with many standard molecular biologylaboratory practices RIPA buffer is one of the recom-mended (or one of the preferred) buffers for efficientrecovery of nuclear proteins [64546]

Protein complexes (Figure 7D) are also slightly more con-centrated in RIPA buffer than in urea Protein complexestend to have high molecular weights and although RIPAbuffer has poor solubility for high molecular weight pro-teins protein complexes are held together largely by non-covalent bonds Detailed Cellular Component DAGs(DAGs not shown) indicate that protein complexesreferred to here include immunoglobulin complexhemoglobin complex fibrinogen complex transcriptorfactor complex DNA polymerase complex DNA-directed

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA)Figure 10Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix proteins and virions are observed almost exclusively in urea fraction

Page 15 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

mitochondrial proteins are preferentially enriched in theRIPA fractions

Selective protein enrichment comparisons based on bothSeqs and Scores are shown in Figures 7 and 8

T2-048T (RIPA) and T2-048T (Urea) fractionsRather than show the entire DAGs cropped views of theextracellular regions of T2-048T (RIPA) and T2-048T(Urea) DAGs (Figure 9) show that extracellular matrixproteins are present almost exclusively in the urea frac-tion It is also seen that mapping found that nearly 90of the original proteins present in the T2-048T proteome(ie 232 of 261 Table 2) have existing Ontologies thusproviding the requisite bioinformatics informationneeded for the characterization of the proteins SimilarlyOntologies are found for 120 of the 143 proteins (84)of the T2-048T proteome

T2-048N1 (RIPA) and T2-048N2 (Urea) fractionsHere again the extracellular regions of T2-048N1 (RIPA)and T2-048N2 (Urea) DAGs (Figure 10) show that extra-cellular matrix proteins are present nearly exclusively inurea buffer with 86 and 80 of the proteins of T2-

048N1 and T2-048N2 proteomes respectively beingfound to have existing Ontologies available It is interest-ing to compare the normalized extracellular matrix Seqsand scores of T2-048T (Tumor) and T2-048N2 (Normal)[T2-048T-UREA (extracellular matrix (Seqs1336Score1680)) T2-048N2-UREA (extracellular matrix(Seqs2959 Score3266))] The normalized scores andSeqs belonging to normal breast tissues are more than twotimes those of the breast tumors which may be due to thedegradation of extracellular matrix proteins in the tumorby proteases to pave the way for the metastasis of cancercells [8-22]

T2-029T1 (RIPA) and T2-029T2 (Urea) fractionsThe cropped-out extracellular regions displayed side-by-side (Figure 11) again show that extracellular matrix pro-teins are found nearly exclusively in urea buffer Andmapping found Ontologies for 84 and 87 of the T2-029T1 and T2-029T2 proteins respectively Upon lower-ing the node filter on the T2-029T2 DAG to zero a fulldisplay of the extracellular matrix proteins are obtained(Additional File 6) Similar lowering of the node filter onthe T2-029T1 DAG did not reveal significantly new or rel-evant structural information

The Cellular Component DAG for the proteome T2-018T (UREA)Figure 4The Cellular Component DAG for the proteome T2-018T (UREA) At a node filter setting of only 13 extracellular matrix proteins are highly evident Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 11 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Complete Cellular Component DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 7 8 9 10 11 and 12 respec-tively

Molecular FunctionMolecular Function annotations also contain elementsthat reveal protein solubilization preferences of the RIPAand urea buffers Thus far Structural Molecule Activity(SMA) is the only functionality in the Molecular Functionannotation that contains nodes that directly relate to theprotein solubility preferences In general the SMA node ofthe urea proteome contains a child node ExtracellularMatrix Structural Constituent which contains thenumber of annotated protein sequences and an annota-tion score for extracellular matrix proteins The SMAnode of the RIPA proteome does not contain any descrip-tors (at the given node filter settings) that suggest the pres-ence of Extracellular Matrix Structural ConstituentThus extracellular matrix proteins are observed nearlyexclusively in the urea fraction

The Molecular Function DAGs for T2-018T (RIPA) andT2-018T (UREA) (Figures 12 and 13 respectively) showclearly that Extracellular Matrix Structural Constituentsare present only in the urea proteome The RIPA and urea

Close-up views of the extracellular regions of T2-018T (UREA)Figure 5Close-up views of the extracellular regions of T2-018T (UREA) clearly show that extracellular matrix proteins dis-solve almost exclusively in urea buffer

Expanded view of the extracellular region of the cellular component DAG for the proteome T2-018T (UREA)Figure 6Expanded view of the extracellular region of the cel-lular component DAG for the proteome T2-018T (UREA) The node filter was reduced to 0 to obtain this complete display Lowering the DAG node filter for the RIPA DAG did not produce appreciable change in the number of nodes displayed within the extracellular region

Page 12 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

DAGs are drawn with node filter settings of 30 and 12respectively

In Figure 14(AndashB) are the cropped-out SMA nodes for T2-018T (RIPA) and T2-018T (UREA) when the node filtersare set to 2 and 1 respectively Further lowering of thenode filters did not produce significant differences in bothDAGs Also shown in Figure 14(CndashD) are the cropped-out SMA nodes for T2-029T1 (RIPA) and T2-029T2(UREA) proteomes Again extracellular matrix proteinsare found nearly exclusively in the urea fractions

The above trend is consistently maintained in Figure 15(A-D) for T2-048T (RIPA) and T2-048T (UREA) and T2-048N1 (RIPA) and T2-048N2 (UREA) respectively

Detailed Molecular Function DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 13 14 15 16 17 and 18respectively

DiscussionThe solubility of proteins in RIPA and urea buffersdepends on several physicochemical factors including the

characteristics of the proteins and properties of the RIPAand urea buffers

Physicochemical properties of the proteins that affecttheir solubility include average charge determined by therelative numbers of Asp Glu Lys and Arg residues andthe content of turn-forming residues (Asn Gly Pro andSer) [39] Insoluble proteins tend to have more hydro-phobic stretches longer than 20 amino acids residueslower glutamine content fewer negatively charged resi-dues and higher percentages of aromatic amino acid resi-dues than soluble ones [40] Indeed high contents ofnegatively charged amino-acid residues and absence ofhydrophobic patches tend to improve protein solubility[41] Also low percentage of aspartic acid glutamic acidasparagines and glutamine residues increases the proba-bility of a protein to be insoluble [41]

Solubility of proteins in lysis buffer also depends highlyon the composition and gross physicochemical propertiesof the lysis buffer These include [4243] the type ofbuffer the presence or absence of phosphate pH saltsampholytes detergents chaotropic agents reducingagents (dithiothreitol (DTT) dithioerythreitol (DTE) β-mercaptoethanol tributyl phosphine (TBP) tris-carboxy-lethylphosphate (TCEP))

Extraction of proteins from breast tumors for proteomic analysisFigure 7Extraction of proteins from breast tumors for pro-teomic analysis nuclear proteins (A) intracellular proteins (C) and protein complexes (D) are more soluble in RIPA buffer than in urea buffer RIPA whereas membrane proteins (B) are slightly more soluble in urea buffer

Extraction of proteins from breast tumors for proteomicsFigure 8Extraction of proteins from breast tumors for pro-teomics proteins of the extracellular region (A) and cytoskeleton (B) are more soluble in urea buffer than in RIPA whereas for cytoplasmic (C) and mitochondrial (D) proteins RIPA buffer is preferred

Page 13 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RIPA and UREA buffers fractionate breast cancer proteins primarily on the basis of molecular weightsRIPA buffer is a versatile and efficient lysis buffer suitablefor the recovery of most proteins including whole cellnuclear mitochondrial membrane receptors cytoskele-tal-associated and soluble proteins However as the datain Figures 2 shows there is a high molecular weight cut-off (ge 12 higher average molecular weight in urea thanRIPA) for a proteins solubility in RIPA buffer Proteins

with molecular weights of around 100 kDa or higher maynot dissolve readily unless they possess unique structuralfeatures that enhance their solubility in RIPA buffer Thuswhen a given protein group is subjected to RIPA bufferextraction the high molecular weight fraction may notdissolve ndash they are recovered as the RIPA-insoluble frac-tion that ultimately dissolve in urea buffer This mayexplain why such a high percentage of proteins are recov-ered in the urea fraction after they have resisted solubility

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA)Figure 9Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) Extracellular matrix proteins and virions are observed almost entirely in urea fraction

Page 14 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

in RIPA buffer (Figures 2D especially Figures 2D (T2-029(TUMOR))

Interestingly Ignatoski and co-workers [42] have alsodemonstrated that different lysis buffers solubilized dif-ferent subsets of cellular proteins (rather than entire pro-teins) based primarily on the molecular weights NeitherRIPA nor urea buffer was however used in this kinaseassay ndash they denature kinases All buffers that they testedwere non-denaturing

Nearly all extracellular matrix proteins are insoluble in RIPA buffer but dissolve readily in urea bufferPerhaps the most important finding in this work is thatnearly all extracellular matrix proteins (ECMs) are insolu-ble in RIPA buffer whereas they dissolve readily in ureabuffer This may be due to ECMs having very high molec-ular weights Why then is RIPA buffer being used rou-tinely to dissolve extracellular matrix proteins byresearchers especially in cancer research Indeed RIPAbuffer does dissolve high molecular weight proteins butthe recovery may be poor The solubility of a protein is acombination of many factors beyond the nature of thelysis buffer If for example the high molecular weightprotein has structural features that enhance its solubility(high contents of negatively charged amino-acid residuesand absence of hydrophobic patches [41]) as discussed inthe introduction section the protein would dissolve inRIPA buffer On the other a smaller molecular weight

protein may surprisingly fail to dissolve in RIPA buffer ifit aggregates or possesses structural features that hamperits solubility Some epigenetic post-translation or spon-taneous structural changes can also impede a proteinssolubility in RIPA buffer One example is tau whichwould normally be soluble in RIPA But when it becomeshyperphosphorylated for example by endogenouslyoverproduced Aβ protein in Alzheimers disease it wouldresist solubility in RIPA buffer [44]

Selective Enrichments of Protein Groups by RIPA and Urea BuffersData in Figure 7A shows that nuclear proteins are some-what more selectively enriched in RIPA buffer than inurea consistent with many standard molecular biologylaboratory practices RIPA buffer is one of the recom-mended (or one of the preferred) buffers for efficientrecovery of nuclear proteins [64546]

Protein complexes (Figure 7D) are also slightly more con-centrated in RIPA buffer than in urea Protein complexestend to have high molecular weights and although RIPAbuffer has poor solubility for high molecular weight pro-teins protein complexes are held together largely by non-covalent bonds Detailed Cellular Component DAGs(DAGs not shown) indicate that protein complexesreferred to here include immunoglobulin complexhemoglobin complex fibrinogen complex transcriptorfactor complex DNA polymerase complex DNA-directed

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA)Figure 10Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix proteins and virions are observed almost exclusively in urea fraction

Page 15 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Complete Cellular Component DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 7 8 9 10 11 and 12 respec-tively

Molecular FunctionMolecular Function annotations also contain elementsthat reveal protein solubilization preferences of the RIPAand urea buffers Thus far Structural Molecule Activity(SMA) is the only functionality in the Molecular Functionannotation that contains nodes that directly relate to theprotein solubility preferences In general the SMA node ofthe urea proteome contains a child node ExtracellularMatrix Structural Constituent which contains thenumber of annotated protein sequences and an annota-tion score for extracellular matrix proteins The SMAnode of the RIPA proteome does not contain any descrip-tors (at the given node filter settings) that suggest the pres-ence of Extracellular Matrix Structural ConstituentThus extracellular matrix proteins are observed nearlyexclusively in the urea fraction

The Molecular Function DAGs for T2-018T (RIPA) andT2-018T (UREA) (Figures 12 and 13 respectively) showclearly that Extracellular Matrix Structural Constituentsare present only in the urea proteome The RIPA and urea

Close-up views of the extracellular regions of T2-018T (UREA)Figure 5Close-up views of the extracellular regions of T2-018T (UREA) clearly show that extracellular matrix proteins dis-solve almost exclusively in urea buffer

Expanded view of the extracellular region of the cellular component DAG for the proteome T2-018T (UREA)Figure 6Expanded view of the extracellular region of the cel-lular component DAG for the proteome T2-018T (UREA) The node filter was reduced to 0 to obtain this complete display Lowering the DAG node filter for the RIPA DAG did not produce appreciable change in the number of nodes displayed within the extracellular region

Page 12 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

DAGs are drawn with node filter settings of 30 and 12respectively

In Figure 14(AndashB) are the cropped-out SMA nodes for T2-018T (RIPA) and T2-018T (UREA) when the node filtersare set to 2 and 1 respectively Further lowering of thenode filters did not produce significant differences in bothDAGs Also shown in Figure 14(CndashD) are the cropped-out SMA nodes for T2-029T1 (RIPA) and T2-029T2(UREA) proteomes Again extracellular matrix proteinsare found nearly exclusively in the urea fractions

The above trend is consistently maintained in Figure 15(A-D) for T2-048T (RIPA) and T2-048T (UREA) and T2-048N1 (RIPA) and T2-048N2 (UREA) respectively

Detailed Molecular Function DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 13 14 15 16 17 and 18respectively

DiscussionThe solubility of proteins in RIPA and urea buffersdepends on several physicochemical factors including the

characteristics of the proteins and properties of the RIPAand urea buffers

Physicochemical properties of the proteins that affecttheir solubility include average charge determined by therelative numbers of Asp Glu Lys and Arg residues andthe content of turn-forming residues (Asn Gly Pro andSer) [39] Insoluble proteins tend to have more hydro-phobic stretches longer than 20 amino acids residueslower glutamine content fewer negatively charged resi-dues and higher percentages of aromatic amino acid resi-dues than soluble ones [40] Indeed high contents ofnegatively charged amino-acid residues and absence ofhydrophobic patches tend to improve protein solubility[41] Also low percentage of aspartic acid glutamic acidasparagines and glutamine residues increases the proba-bility of a protein to be insoluble [41]

Solubility of proteins in lysis buffer also depends highlyon the composition and gross physicochemical propertiesof the lysis buffer These include [4243] the type ofbuffer the presence or absence of phosphate pH saltsampholytes detergents chaotropic agents reducingagents (dithiothreitol (DTT) dithioerythreitol (DTE) β-mercaptoethanol tributyl phosphine (TBP) tris-carboxy-lethylphosphate (TCEP))

Extraction of proteins from breast tumors for proteomic analysisFigure 7Extraction of proteins from breast tumors for pro-teomic analysis nuclear proteins (A) intracellular proteins (C) and protein complexes (D) are more soluble in RIPA buffer than in urea buffer RIPA whereas membrane proteins (B) are slightly more soluble in urea buffer

Extraction of proteins from breast tumors for proteomicsFigure 8Extraction of proteins from breast tumors for pro-teomics proteins of the extracellular region (A) and cytoskeleton (B) are more soluble in urea buffer than in RIPA whereas for cytoplasmic (C) and mitochondrial (D) proteins RIPA buffer is preferred

Page 13 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RIPA and UREA buffers fractionate breast cancer proteins primarily on the basis of molecular weightsRIPA buffer is a versatile and efficient lysis buffer suitablefor the recovery of most proteins including whole cellnuclear mitochondrial membrane receptors cytoskele-tal-associated and soluble proteins However as the datain Figures 2 shows there is a high molecular weight cut-off (ge 12 higher average molecular weight in urea thanRIPA) for a proteins solubility in RIPA buffer Proteins

with molecular weights of around 100 kDa or higher maynot dissolve readily unless they possess unique structuralfeatures that enhance their solubility in RIPA buffer Thuswhen a given protein group is subjected to RIPA bufferextraction the high molecular weight fraction may notdissolve ndash they are recovered as the RIPA-insoluble frac-tion that ultimately dissolve in urea buffer This mayexplain why such a high percentage of proteins are recov-ered in the urea fraction after they have resisted solubility

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA)Figure 9Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) Extracellular matrix proteins and virions are observed almost entirely in urea fraction

Page 14 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

in RIPA buffer (Figures 2D especially Figures 2D (T2-029(TUMOR))

Interestingly Ignatoski and co-workers [42] have alsodemonstrated that different lysis buffers solubilized dif-ferent subsets of cellular proteins (rather than entire pro-teins) based primarily on the molecular weights NeitherRIPA nor urea buffer was however used in this kinaseassay ndash they denature kinases All buffers that they testedwere non-denaturing

Nearly all extracellular matrix proteins are insoluble in RIPA buffer but dissolve readily in urea bufferPerhaps the most important finding in this work is thatnearly all extracellular matrix proteins (ECMs) are insolu-ble in RIPA buffer whereas they dissolve readily in ureabuffer This may be due to ECMs having very high molec-ular weights Why then is RIPA buffer being used rou-tinely to dissolve extracellular matrix proteins byresearchers especially in cancer research Indeed RIPAbuffer does dissolve high molecular weight proteins butthe recovery may be poor The solubility of a protein is acombination of many factors beyond the nature of thelysis buffer If for example the high molecular weightprotein has structural features that enhance its solubility(high contents of negatively charged amino-acid residuesand absence of hydrophobic patches [41]) as discussed inthe introduction section the protein would dissolve inRIPA buffer On the other a smaller molecular weight

protein may surprisingly fail to dissolve in RIPA buffer ifit aggregates or possesses structural features that hamperits solubility Some epigenetic post-translation or spon-taneous structural changes can also impede a proteinssolubility in RIPA buffer One example is tau whichwould normally be soluble in RIPA But when it becomeshyperphosphorylated for example by endogenouslyoverproduced Aβ protein in Alzheimers disease it wouldresist solubility in RIPA buffer [44]

Selective Enrichments of Protein Groups by RIPA and Urea BuffersData in Figure 7A shows that nuclear proteins are some-what more selectively enriched in RIPA buffer than inurea consistent with many standard molecular biologylaboratory practices RIPA buffer is one of the recom-mended (or one of the preferred) buffers for efficientrecovery of nuclear proteins [64546]

Protein complexes (Figure 7D) are also slightly more con-centrated in RIPA buffer than in urea Protein complexestend to have high molecular weights and although RIPAbuffer has poor solubility for high molecular weight pro-teins protein complexes are held together largely by non-covalent bonds Detailed Cellular Component DAGs(DAGs not shown) indicate that protein complexesreferred to here include immunoglobulin complexhemoglobin complex fibrinogen complex transcriptorfactor complex DNA polymerase complex DNA-directed

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA)Figure 10Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix proteins and virions are observed almost exclusively in urea fraction

Page 15 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

DAGs are drawn with node filter settings of 30 and 12respectively

In Figure 14(AndashB) are the cropped-out SMA nodes for T2-018T (RIPA) and T2-018T (UREA) when the node filtersare set to 2 and 1 respectively Further lowering of thenode filters did not produce significant differences in bothDAGs Also shown in Figure 14(CndashD) are the cropped-out SMA nodes for T2-029T1 (RIPA) and T2-029T2(UREA) proteomes Again extracellular matrix proteinsare found nearly exclusively in the urea fractions

The above trend is consistently maintained in Figure 15(A-D) for T2-048T (RIPA) and T2-048T (UREA) and T2-048N1 (RIPA) and T2-048N2 (UREA) respectively

Detailed Molecular Function DAGs for T2-048T (RIPA)and T2-048T (Urea) T2-048N1 (RIPA) and T2-048N2(Urea) and T2-029T1 (RIPA) and T2-029T2 (Urea) areavailable as Additional Files 13 14 15 16 17 and 18respectively

DiscussionThe solubility of proteins in RIPA and urea buffersdepends on several physicochemical factors including the

characteristics of the proteins and properties of the RIPAand urea buffers

Physicochemical properties of the proteins that affecttheir solubility include average charge determined by therelative numbers of Asp Glu Lys and Arg residues andthe content of turn-forming residues (Asn Gly Pro andSer) [39] Insoluble proteins tend to have more hydro-phobic stretches longer than 20 amino acids residueslower glutamine content fewer negatively charged resi-dues and higher percentages of aromatic amino acid resi-dues than soluble ones [40] Indeed high contents ofnegatively charged amino-acid residues and absence ofhydrophobic patches tend to improve protein solubility[41] Also low percentage of aspartic acid glutamic acidasparagines and glutamine residues increases the proba-bility of a protein to be insoluble [41]

Solubility of proteins in lysis buffer also depends highlyon the composition and gross physicochemical propertiesof the lysis buffer These include [4243] the type ofbuffer the presence or absence of phosphate pH saltsampholytes detergents chaotropic agents reducingagents (dithiothreitol (DTT) dithioerythreitol (DTE) β-mercaptoethanol tributyl phosphine (TBP) tris-carboxy-lethylphosphate (TCEP))

Extraction of proteins from breast tumors for proteomic analysisFigure 7Extraction of proteins from breast tumors for pro-teomic analysis nuclear proteins (A) intracellular proteins (C) and protein complexes (D) are more soluble in RIPA buffer than in urea buffer RIPA whereas membrane proteins (B) are slightly more soluble in urea buffer

Extraction of proteins from breast tumors for proteomicsFigure 8Extraction of proteins from breast tumors for pro-teomics proteins of the extracellular region (A) and cytoskeleton (B) are more soluble in urea buffer than in RIPA whereas for cytoplasmic (C) and mitochondrial (D) proteins RIPA buffer is preferred

Page 13 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RIPA and UREA buffers fractionate breast cancer proteins primarily on the basis of molecular weightsRIPA buffer is a versatile and efficient lysis buffer suitablefor the recovery of most proteins including whole cellnuclear mitochondrial membrane receptors cytoskele-tal-associated and soluble proteins However as the datain Figures 2 shows there is a high molecular weight cut-off (ge 12 higher average molecular weight in urea thanRIPA) for a proteins solubility in RIPA buffer Proteins

with molecular weights of around 100 kDa or higher maynot dissolve readily unless they possess unique structuralfeatures that enhance their solubility in RIPA buffer Thuswhen a given protein group is subjected to RIPA bufferextraction the high molecular weight fraction may notdissolve ndash they are recovered as the RIPA-insoluble frac-tion that ultimately dissolve in urea buffer This mayexplain why such a high percentage of proteins are recov-ered in the urea fraction after they have resisted solubility

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA)Figure 9Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) Extracellular matrix proteins and virions are observed almost entirely in urea fraction

Page 14 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

in RIPA buffer (Figures 2D especially Figures 2D (T2-029(TUMOR))

Interestingly Ignatoski and co-workers [42] have alsodemonstrated that different lysis buffers solubilized dif-ferent subsets of cellular proteins (rather than entire pro-teins) based primarily on the molecular weights NeitherRIPA nor urea buffer was however used in this kinaseassay ndash they denature kinases All buffers that they testedwere non-denaturing

Nearly all extracellular matrix proteins are insoluble in RIPA buffer but dissolve readily in urea bufferPerhaps the most important finding in this work is thatnearly all extracellular matrix proteins (ECMs) are insolu-ble in RIPA buffer whereas they dissolve readily in ureabuffer This may be due to ECMs having very high molec-ular weights Why then is RIPA buffer being used rou-tinely to dissolve extracellular matrix proteins byresearchers especially in cancer research Indeed RIPAbuffer does dissolve high molecular weight proteins butthe recovery may be poor The solubility of a protein is acombination of many factors beyond the nature of thelysis buffer If for example the high molecular weightprotein has structural features that enhance its solubility(high contents of negatively charged amino-acid residuesand absence of hydrophobic patches [41]) as discussed inthe introduction section the protein would dissolve inRIPA buffer On the other a smaller molecular weight

protein may surprisingly fail to dissolve in RIPA buffer ifit aggregates or possesses structural features that hamperits solubility Some epigenetic post-translation or spon-taneous structural changes can also impede a proteinssolubility in RIPA buffer One example is tau whichwould normally be soluble in RIPA But when it becomeshyperphosphorylated for example by endogenouslyoverproduced Aβ protein in Alzheimers disease it wouldresist solubility in RIPA buffer [44]

Selective Enrichments of Protein Groups by RIPA and Urea BuffersData in Figure 7A shows that nuclear proteins are some-what more selectively enriched in RIPA buffer than inurea consistent with many standard molecular biologylaboratory practices RIPA buffer is one of the recom-mended (or one of the preferred) buffers for efficientrecovery of nuclear proteins [64546]

Protein complexes (Figure 7D) are also slightly more con-centrated in RIPA buffer than in urea Protein complexestend to have high molecular weights and although RIPAbuffer has poor solubility for high molecular weight pro-teins protein complexes are held together largely by non-covalent bonds Detailed Cellular Component DAGs(DAGs not shown) indicate that protein complexesreferred to here include immunoglobulin complexhemoglobin complex fibrinogen complex transcriptorfactor complex DNA polymerase complex DNA-directed

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA)Figure 10Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix proteins and virions are observed almost exclusively in urea fraction

Page 15 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RIPA and UREA buffers fractionate breast cancer proteins primarily on the basis of molecular weightsRIPA buffer is a versatile and efficient lysis buffer suitablefor the recovery of most proteins including whole cellnuclear mitochondrial membrane receptors cytoskele-tal-associated and soluble proteins However as the datain Figures 2 shows there is a high molecular weight cut-off (ge 12 higher average molecular weight in urea thanRIPA) for a proteins solubility in RIPA buffer Proteins

with molecular weights of around 100 kDa or higher maynot dissolve readily unless they possess unique structuralfeatures that enhance their solubility in RIPA buffer Thuswhen a given protein group is subjected to RIPA bufferextraction the high molecular weight fraction may notdissolve ndash they are recovered as the RIPA-insoluble frac-tion that ultimately dissolve in urea buffer This mayexplain why such a high percentage of proteins are recov-ered in the urea fraction after they have resisted solubility

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA)Figure 9Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) Extracellular matrix proteins and virions are observed almost entirely in urea fraction

Page 14 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

in RIPA buffer (Figures 2D especially Figures 2D (T2-029(TUMOR))

Interestingly Ignatoski and co-workers [42] have alsodemonstrated that different lysis buffers solubilized dif-ferent subsets of cellular proteins (rather than entire pro-teins) based primarily on the molecular weights NeitherRIPA nor urea buffer was however used in this kinaseassay ndash they denature kinases All buffers that they testedwere non-denaturing

Nearly all extracellular matrix proteins are insoluble in RIPA buffer but dissolve readily in urea bufferPerhaps the most important finding in this work is thatnearly all extracellular matrix proteins (ECMs) are insolu-ble in RIPA buffer whereas they dissolve readily in ureabuffer This may be due to ECMs having very high molec-ular weights Why then is RIPA buffer being used rou-tinely to dissolve extracellular matrix proteins byresearchers especially in cancer research Indeed RIPAbuffer does dissolve high molecular weight proteins butthe recovery may be poor The solubility of a protein is acombination of many factors beyond the nature of thelysis buffer If for example the high molecular weightprotein has structural features that enhance its solubility(high contents of negatively charged amino-acid residuesand absence of hydrophobic patches [41]) as discussed inthe introduction section the protein would dissolve inRIPA buffer On the other a smaller molecular weight

protein may surprisingly fail to dissolve in RIPA buffer ifit aggregates or possesses structural features that hamperits solubility Some epigenetic post-translation or spon-taneous structural changes can also impede a proteinssolubility in RIPA buffer One example is tau whichwould normally be soluble in RIPA But when it becomeshyperphosphorylated for example by endogenouslyoverproduced Aβ protein in Alzheimers disease it wouldresist solubility in RIPA buffer [44]

Selective Enrichments of Protein Groups by RIPA and Urea BuffersData in Figure 7A shows that nuclear proteins are some-what more selectively enriched in RIPA buffer than inurea consistent with many standard molecular biologylaboratory practices RIPA buffer is one of the recom-mended (or one of the preferred) buffers for efficientrecovery of nuclear proteins [64546]

Protein complexes (Figure 7D) are also slightly more con-centrated in RIPA buffer than in urea Protein complexestend to have high molecular weights and although RIPAbuffer has poor solubility for high molecular weight pro-teins protein complexes are held together largely by non-covalent bonds Detailed Cellular Component DAGs(DAGs not shown) indicate that protein complexesreferred to here include immunoglobulin complexhemoglobin complex fibrinogen complex transcriptorfactor complex DNA polymerase complex DNA-directed

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA)Figure 10Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix proteins and virions are observed almost exclusively in urea fraction

Page 15 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

in RIPA buffer (Figures 2D especially Figures 2D (T2-029(TUMOR))

Interestingly Ignatoski and co-workers [42] have alsodemonstrated that different lysis buffers solubilized dif-ferent subsets of cellular proteins (rather than entire pro-teins) based primarily on the molecular weights NeitherRIPA nor urea buffer was however used in this kinaseassay ndash they denature kinases All buffers that they testedwere non-denaturing

Nearly all extracellular matrix proteins are insoluble in RIPA buffer but dissolve readily in urea bufferPerhaps the most important finding in this work is thatnearly all extracellular matrix proteins (ECMs) are insolu-ble in RIPA buffer whereas they dissolve readily in ureabuffer This may be due to ECMs having very high molec-ular weights Why then is RIPA buffer being used rou-tinely to dissolve extracellular matrix proteins byresearchers especially in cancer research Indeed RIPAbuffer does dissolve high molecular weight proteins butthe recovery may be poor The solubility of a protein is acombination of many factors beyond the nature of thelysis buffer If for example the high molecular weightprotein has structural features that enhance its solubility(high contents of negatively charged amino-acid residuesand absence of hydrophobic patches [41]) as discussed inthe introduction section the protein would dissolve inRIPA buffer On the other a smaller molecular weight

protein may surprisingly fail to dissolve in RIPA buffer ifit aggregates or possesses structural features that hamperits solubility Some epigenetic post-translation or spon-taneous structural changes can also impede a proteinssolubility in RIPA buffer One example is tau whichwould normally be soluble in RIPA But when it becomeshyperphosphorylated for example by endogenouslyoverproduced Aβ protein in Alzheimers disease it wouldresist solubility in RIPA buffer [44]

Selective Enrichments of Protein Groups by RIPA and Urea BuffersData in Figure 7A shows that nuclear proteins are some-what more selectively enriched in RIPA buffer than inurea consistent with many standard molecular biologylaboratory practices RIPA buffer is one of the recom-mended (or one of the preferred) buffers for efficientrecovery of nuclear proteins [64546]

Protein complexes (Figure 7D) are also slightly more con-centrated in RIPA buffer than in urea Protein complexestend to have high molecular weights and although RIPAbuffer has poor solubility for high molecular weight pro-teins protein complexes are held together largely by non-covalent bonds Detailed Cellular Component DAGs(DAGs not shown) indicate that protein complexesreferred to here include immunoglobulin complexhemoglobin complex fibrinogen complex transcriptorfactor complex DNA polymerase complex DNA-directed

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA)Figure 10Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix proteins and virions are observed almost exclusively in urea fraction

Page 15 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

RNA polymerase II holoenzyme RNA polymerase com-plex nucleosome myosin laminin complex membraneattack complex mediator complex tubulin MHC proteincomplex and ribonucleocomplex

Mitochondrial proteins also appear to be slightly morefavored by RIPA buffer than urea (Figure 8D) Again RIPAbuffer is one of the recommended lysis buffers for therecovery of mitochondrial proteins for Western blot [6]

On the other hand urea buffer is clearly more efficient inselectively enriching extracellular region and cytoskeletalproteins (Figures 7A and 7B) in addition to extracellularmatrix proteins already discussed Neither RIPA buffer norurea buffer is significantly preponderant in selectiveenrichment of membrane intracellular or cytoplasmicproteins (Figures 7B and 7C and Figure 8C respectively)

Limitations of RIPA and Urea BuffersThere is no single lysis buffer that would solubilize allclasses of proteins however Each buffer has its pros and

cons Some of the known limitations of RIPA and ureabuffers are highlighted below

RIPA buffer-induced post-lysis modulation of biochemical pathwaysRIPA buffer has been shown to alter some biochemicalpathways leading to experimental results that may bespurious Hence the need to verify data by using otherlysis buffers Two examples are provided herein

In one example DeSeau and co-workers [47] showed thatthe level of pp60c-src kinase activity detected in immunecomplex protein kinase assays can be substantially modu-lated by RIPA buffer They thus advise that comparing ofthe results of pp60c-src in vitro protein kinase assays inother cellular systems where only RIPA buffer lysis hasbeen used should be interpreted with caution Specifi-cally they found that the in vitro protein kinase activity ofpp60c-src molecules derived from RIPA buffer lysates ofcolon carcinoma cells was elevated five- to sevenfoldwhen compared with pp60c-src from the same cells lysed in

Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA)Figure 11Close-up views of the extracellular regions of the Cellular Component DAG of the proteomes T2-029T (RIPA) and T2-029T (UREA) Extracellular matrix proteins are observed almost exclusivly in urea fraction (right)

Page 16 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

a buffer containing only Nonidet-P 40 Additionally theyfound that in RIPA buffer the difference in specific activ-ity of pp60c-src between normal colon mucosal cells andcolon carcinoma cells is about ten-to thirtyfold whereaswith a lysis buffer containing only Nonidet-P 40 as adetergent the difference would be less than three- to four-fold Thus if Nonidet-P 40 or other lysis buffers were notused in an effort to validate data obtained in RIPA bufferthe entire data on this work could have been in error

In another example abnormally high caspase-3 and -7activity in stimulated human peripheral blood lym-phocytes (PBLs) has been shown to be a spurious sideeffect caused by RIPA buffer that was used to lyse the acti-vated T-lymphocytes [48] In contrast when a lysis buffercontaining 2 SDS was used the caspases remained intheir zymogen proforms and no proteolytic processing ofcaspase substrates was detected It was subsequentlydetermined that the release liberation of GraB or similarproteases from cytotoxic granules during the lysis proce-dure was responsible for artifactual activation of caspase-3 RIPA may disrupt GraB-containing granules more effi-ciently than 02 Nonidet P-40 or other lysis buffers used[48]

Many protein groups are insoluble in urea bufferAlthough urea buffer has proven very effective in dissolv-ing extracellular matrix proteins and a wide range of otherprotein groups it nevertheless has limitations In fact

Granier [49] noted that many membrane proteins areinsoluble in urea if extracted without heating And asmentioned in the Background section above heating ureain the presence of proteins most likely would result in cov-alent modifications of the protein by the hydrolysis prod-ucts produced by heating urea (eg carbamylation ofproteins by isocyanate) Thus urea does not possess uni-versal solubility for all membrane proteins Ames andNikaido [50] solubilized membrane proteins of salmonellatyphimurium with hot SDS when even the most powerfulOFarrells buffer (urea buffer) [51] failed to dissolve themembrane proteins

A cell surface proteoglycan with a molecular weight of450 kDa was also found to be very insoluble in ureabuffer [16]

In general many proteins that have proven insoluble inurea buffer are shown to be human lens proteins [5253]especially cataractous proteins [52-54] Weber andMcFadden described a heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells Theyfound that about 5 of the total cellular proteins synthe-sized in exponentially dividing PC12 pheochromocytomacells remained insoluble even in 6 M urea [55]

A major factor that decreases the solubility of proteins inurea is the formation of disulfide cross-bridges which canbe acquired by a protein through aerobic oxidation of

Molecular Function DAG for the proteome T2-018T (RIPA)Figure 12Molecular Function DAG for the proteome T2-018T (RIPA) Extracellular matrix structural constituents are not seen even at a node filter setting of 12

Page 17 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

thiol groups Even a small molecular weight protein couldbecome very insoluble in urea upon formation of disul-phide bridges This was the case with a 42 kDa Rec12(Spo11) meiotic recombinase of fission yeast (Rec12 pro-tein) [56] that was expressed in E coli Rec12 proteinresisted solubility in 6 M urea but was ultimatelyextracted with 6 M Guanidine hydrochloride [56] Subse-quent analyses showed that it has four disulfide bridgesthat impeded its solubility in urea Human eye lens pro-teins acquire disulfide cross-bridges by exposure to hyper-baric oxygen (Reviewed in Ref [52]) The eye lens proteinsthen become opaque cataractous and resist solubility inurea buffer [52]

Another example is the human centrosomal proteinwhich exists as a doublet of 6264 kDa and is insoluble ineven 8 M urea (a condition that would dissolve mostknown centrosomal proteins) [57]

Preferential solubilization of extracellular matrix proteins in urea lysis buffer variablesDespite differences in the breast tumors analyzed in thiswork (Table 1) a common feature remains the preferen-tial solubilization of extracellular matrix proteins in urealysis buffer Differences include (Table 1) breast location(right breast left breast left breast bilateral) Diagnosis

(Infiltrating Ductal Carcinoma Infiltrating Ductal Carci-noma Adenocarcinoma) Stage of the tumor (IVT4N0M0 IIB T2N1M0 IIB T4N1M0) Grade of the tumor(II I I III) Patient Age (yrs 75 39 39 47) and grossfindings)

ConclusionThis work shows that most extracellular matrix proteins(ECM) in the breast tumors and matched normal tissuesin this work are dissolved in the urea buffer fraction theyare mostly insoluble in RIPA buffer Because ECMs arehighly important in cancer including tumor develop-ment progression adhesion and metastasis importantinformation may be missed in cancer research if they arenot efficiently extracted by RIPA buffer

This work also shows that RIPA and urea lysis buffers frac-tionate tissue proteins primarily on the basis of molecularweights The average molecular weight of proteins thatdissolve exclusively in urea buffer is higher (up to 60)than in RIPA

Protein complexes nuclear mitochondrial cytoplasmicand intracellular proteins are more soluble in RIPA bufferthan in urea whereas membrane cytoskeletal and extra-cellular region proteins are more soluble in urea buffer

Molecular Function DAG for the proteome T2-018T (UREA)Figure 13Molecular Function DAG for the proteome T2-018T (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellular matrix structural constituents None is observed in the RIPA buffer fraction DAG of Figure 12 shown above Thus extracellular matrix proteins are soluble primarily in urea buffer

Page 18 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

For proteomic analyses of breast tumors and other solidtumors a two-step extraction process is herein recom-mended First the tumor should be subjected to RIPA pro-tein extraction Second the insoluble matter left afterRIPA extraction should be probed for residual proteincontent by extracting with the urea-based buffer describedin this work

Competing interestsThe authors declare that they have no competing interests

Authors contributionsLCN carried out all of this work using breast tumor andnormal breast tissue lysates provide by Protein Biotech-nologies Inc (Ramona VA) Mr Phillip E SchwartzPresident and CEO of Protein Biotechnologies Inc isacknowledged for his generous gift of extra free breastlysates after an initial purchase from his company Allauthors read and approved the final manuscript

Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel BFigure 14Close-up views of the Structural Molecule Activity node of T2-018T (RIPA) Panel A compared to the SMA of T2-018T (UREA) Panel B The SMA of T2-018T (UREA) contains a node corresponding to extracellular matrix structural constituents whereas no ECM constituents are seen in the SMA of T2-018T (RIPA) Panels C and D compare the SMA nodes of T2-029T (RIPA) and T2-029T (UREA) respectively wherein it is seen that no extracellular matrix structural constituents are present in T2-029T (RIPA)

Page 19 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Page 20 of 24(page number not for citation purposes)

Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA)Figure 15Close-up views of the Structural Molecule Activity of the Molecular Function DAG of the proteomes T2-048T (RIPA) and T2-048T (UREA) and T2-048N (RIPA) and T2-048N (UREA) Extracellular matrix structural constitu-ents are observed nearly exclusively in the urea fractions Panels B and D

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Additional material

Additional file 1BLAST Table for discoidin domain receptor 1 (DDR1) BLAST Table for discoidin domain receptor 1 (gi|68533097|dbj|BAE061031|894) For each protein input query the BLASTmachine generates a BLAST Table [34] The table shows parameters of the BLAST search including Sequences producing significant alignments Gene Name Acces-sion number e-Value align-length positives similarity hsp and mapping (Ontologies found) for each of the 40 hits requestedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S1xls]

Additional file 2MudPIT Mass Spectra of the breast tumor T2-018 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at right A typical MudPIT experiment consists of a 12-cycle run in which a 60-minute nano-LC gradient is run for each of 1 1D_2 μL sample 2 2D_10 μL sample 3 2D_0 mM NH4COO- 4 2D_25 mM NH4COO- 5 2D_50 mM NH4COO- 6 2D_75 mM NH4COO- 7 2D_100 mM NH4COO-

8 2D_150 mM NH4COO- 9 2D_200 mM NH4COO- 10 2D_250 mM NH4COO- 11 2D_300 mM NH4COO- and 12 2D_500 mM NH4COO- NH4COO- is ammonium formateClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S2tiff]

Additional file 3MudPIT Mass Spectra of the breast tumor T2-048 TUMOR MudPIT Mass Spectra of the breast tumor T2-048 TUMOR Spectra of RIPA-sol-uble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S3tiff]

Additional file 4MudPIT Mass Spectra of the matched normal breast tissue T2-048 NORMAL The set of 12 MudPIT mass spectra of the RIPA-soluble frac-tion are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S4tiff]

Additional file 5MudPIT Mass Spectra of the bilateral breast tumor T2-029 TUMOR The set of 12 MudPIT mass spectra of the RIPA-soluble fraction are shown at left whereas those for the urea-soluble fraction are shown at rightClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S5tiff]

Additional file 6Expanded view of the extracellular region of the Cellular Component DAG for the bilateral proteome T2-029T (UREA) The node filter was reduced to 0 to obtain this complete display In contrast lowering the DAG node filter for the RIPA DAG counterpart did not produce appreci-able change in the number of nodes displayed within the extracellular regionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S6png]

Additional file 7Cellular Component DAG for the proteome T2-048T (RIPA) Extra-cellular matrix proteins are not seen even when the node filter is lowered to 10Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S7png]

Additional file 8Cellular Component DAG for the proteome T2-048T (UREA) At a node filter setting of 16 there are 31 extracellular matrix proteins Thus extracellular matrix proteins are soluble almost exclusively in urea buffer None is seen in the RIPA buffer fraction of this proteome (Additional File 7)Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S8png]

Additional file 9Cellular Component DAG of the matched normal proteome T2-048N (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 14Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S9png]

Additional file 10Cellular Component DAG of the matched normal proteome T2-048N (UREA) Twenty-nine extracellular matrix proteins are present in this urea buffer fraction of T2-048N even when the DAG is displayed with a node filter setting of just 5 (cf Node filter is 14 in Additional File 9 above) no extracellular matrix proteins are present in the RIPA buffer fraction of this proteome shown in Additional File 9 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S10png]

Additional file 11Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (RIPA) Extracellular matrix proteins are not seen even at a node filter setting of 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S11png]

Page 21 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

AcknowledgementsBreast tumor and normal breast tissue lysates were obtained from Protein Biotechnologies Inc (Ramona VA) Many thanks to Mr Phillip E Schwartz President and CEO of Protein Biotechnologies Inc for his generous gift of extra free breast tumor lysates after an initial purchase from his company Thanks also to John B Mangrum (Brad) for his encouragements assistance and helpful suggestions

Data are generated at the Virginia Commonwealth University Mass Spec-trometry Resource for the Study of Biological Complexity httpwwwvcueducsbcmsrsbc Support for this facility from the National Sci-ence Foundation VCU Life Sciences the VCU Center for the Study of Bio-logical Complexity the Massey Cancer Center the Department of Internal Medicine and the School of Medicine the Higher Education Equipment Trust Fund of the State of Virginia the College of Humanities and Sciences and other academic units at VCU is gratefully acknowledged

References1 Rabilloud T Solubilization of proteins for electrophoretic

analyses Electrophoresis 1996 17813-8292 Harlow E Lane D Antibodies A Laboratory Manual Cold Spring Harbor

NY Cold Spring Harbor Press 1988 3 Molloy MP Herbert BR Walsh BJ Tyler MI Traini M Sanchez J-C

Hochstrasser DF Williams KL Gooley AA Extraction of mem-brane proteins by differential solubilization for separationusing two-dimensional gel electrophoresis Electrophoresis1998 19837-844

4 Rabilloud T Adessi C Giraudel A Lunardi J Improvement of thesolubilization of proteins in two-dimensional electrophoresiswith immobilized pH gradients Electrophoresis 199718307-316

5 Poirier F Lawrence D Vigier P Jullien P A ts T mutant of SchmidtRuppin strain of Rous sarcoma virus restricted at 395 Deg Cfor the morphological transformation and the tumorigenic-ity of chicken embryo fibroblasts Int J Cancer 1982 2969-76

6 Abcam Western blotting ndash a beginners guide httpwwwabcamcomtechnical httpwwwabcamcompspdfprotocolsWB-beginnerpdf Retrieved on October 7 2008

7 Singh S Powell DW Rane MJ Millard TH Trent JO Pierce WM KleinJB Machesky LM McLeish KR Identification of the p16-arc sub-unit of the arp 23 complex as a substrate of MAPK-activatedprotein kinase 2 by proteomic analysis J Biol Chem 200327836410-36417

8 Wang F Gao J Relationship between extracellular matrix andprogressive growth of malignant tumor Zhonghua ZhongliuZazhi 1998 20112-115

9 Veeck J Chorovicer M Naami A Breuer E Zafrakas M Bektas NDurst M Kristiansen G Wild P Hartmann A et al The extracellu-

Additional file 12Cellular Component DAG of the bilateral Adenocarcinoma proteome T2-029T (UREA) Thirty-seven extracellular matrix proteins are observed at a node filter setting of 16 No extracellular matrix proteins are observed in the RIPA buffer fraction of this proteome which is shown in Additional File 11 aboveClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S12png]

Additional file 13Molecular Function DAG for the proteome T2-048T (RIPA) Extracel-lular matrix structural constituents are not seen even when the node filter is set at 12Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S13png]

Additional file 14Molecular Function DAG of the proteome T2-048T (UREA) The Structural Molecule Activity (SMA) of the urea proteome contains 20 extracellular matrix structural constituents none of which is observed in the RIPA buffer fraction DAG of Additional File 13 shown above Thus extracellular matrix proteins are soluble primarily in urea bufferClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S14png]

Additional file 15Molecular Function DAG of the matched normal proteome in RIPA buffer T2-048N (RIPA) Extracellular matrix structural constituents are not seen even when the node filter is set at 17Click here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S15png]

Additional file 16Molecular Function DAG for the proteome T2-048N (UREA) The Structural Molecule Activity of the urea proteome contains 13 extracellu-lar matrix structural constituents None is observed in the RIPA buffer fraction DAG of Additional File 15 shown above Thus extracellular matrix proteins appear to be soluble primarily in urea bufferClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S16png]

Additional file 17Molecular Function DAG of the matched normal proteome T2-029T (RIPA) Extracellular matrix structural constituents are not observed even at a node filter setting of 19 Thus extracellular matrix proteins do not appear to be soluble in RIPA bufferClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S17png]

Additional file 18Molecular Function DAG for the proteome T2-029T (UREA) The Structural Molecule Activity of the urea proteome contains 25 extracellu-lar matrix structural constituents None of these constituents is observed in the RIPA buffer fraction DAG of Additional File 17 shown above Thus extracellular matrix proteins are soluble primarily in urea bufferClick here for file[httpwwwbiomedcentralcomcontentsupplementary1477-5956-6-30-S18png]

Page 22 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

lar matrix protein ITIH5 is a novel prognostic marker ininvasive node-negative breast cancer and its aberrantexpression is caused by promoter hypermethylation Onco-gene 2007 27(6)865-876

10 Scott GK Proteinases and proteinase inhibitors in tumor cellgrowth and metastasis Cancer Journal 1997 1080-86

11 Schmidt R Buumlltmann A Ungerer M Joghetaei N Buumllbuumll Ouml Thieme SChavakis T Toole BP Gawaz M Schoumlmig A May AE Extracellularmatrix metalloproteinase inducer regulates matrix metallo-proteinase activity in cardiovascular cells implications inacute myocardial infarction Circulation 2006 113834-841

12 Ruoslahti E The Walter Herbert Lecture Control of cellmotility and tumour invasion by extracellular matrix inter-actions Brit J Cancer 1992 66239-242

13 Ortega-Velazquez R Gonzalez-Rubio M Ruiz-Torres MP Diez-Marques ML Iglesias MC Rodriacuteguez-Puyol M D R-P Collagen Iupregulates extracellular matrix gene expression and secre-tion of TGF-beta1 by cultured human mesangial cells Am JPhysiol Cell Physiol 2004 2861335-1343

14 Obrist P Spizzo G Ensinger C Fong D Brunhuber T Schaumlfer GVarga M Margreiter R Amberger A Gastl G Christiansen M Aber-rant tetranectin expression in human breast carcinomas asa predictor of survival J Clin Pathol 2004 57417-421

15 Midwood KS Schwarzbauer JE Tenascin-C modulates matrixcontraction via focal adhesion kinase ndash and Rho-mediatedsignaling pathways Mol Biol Cell 2002 133601-3613

16 Mercurius KO Morla AO Cell adhesion and signaling on thefibronectin 1st type III repeat requisite roles for cell surfaceproteoglycans and integrins BMC Cell Biology 2001 2

17 Matrisian LM Wright J Newell K Witty JP Matrix-degrading met-alloproteinases in tumor progression Princess Takamatsu Symp1994 24152-161

18 Macura-Biegun A The role of extracellular matrix in tumorprogression Central-Eur J Immunol 1998 238-14

19 Kelly T Yan Y Osborne RL Athota AB Rozypal TL Colclasure JCChu WS Proteolysis of extracellular matrix by Invadopodiafacilitates human breast cancer cell invasion and is mediatedby matrix metalloproteinases Clin Experimental Metastasis 199816501-512

20 Klees RF Salasznyk RM Vandenberg S Bennett K Plopper GE Lam-inin-5 activates extracellular matrix production and osteo-genic gene focusing in human mesenchymal stem cellsMatrix Biol 2007 26106-114

21 Kobayashi H Mechanism of tumor cell-induced extracellularmatrix degradation inhibition of cell-surface proteolyticactivity might have a therapeutic effect on tumor cell inva-sion and metastasis Nippon Sanka Fujinka Gakkai Zasshi 199648(8)623-632

22 Hornebeck W Maquart FX Proteolyzed matrix as a templatefor the regulation of tumor progression Biomedicine amp Pharma-cotherapy 2003 57223-230

23 Fontana S Pucci-Minafra I Becchi M Freyria A-M Minafra S Effectof collagen substrates on proteomic modulation of breastcancer cells Proteomics 2004 4849-860

24 Fata JE Werb Z Bissell MJ Regulation of mammary gland mor-phogenesis by the extracellular matrix and its remodelingenzymes Breast Cancer Res 2004 61-11

25 Benvenuti S Cramer R Quinn CC Bruce J Zvelebil M Corless SBond J Yang A Hockfield S Burlingame AL et al Differential pro-teome analysis of replicative senescence in rat embryofibroblasts Mol Cellular Proteomics 2002 1280-292

26 Espan L Martin B Aragues R Chiva C Oliva B Andreu D Sierra ABcl-xL-mediated changes in metabolic pathways of breastcancer cells Am J Pathol 2005 1671125-1137

27 Jung SY Malovannaya A Wei J OMalley BW Qin J Proteomicanalysis of steady-state nuclear hormone receptor coactiva-tor complexes Molecular Endocrinol 2005 192451-2465

28 Pucci-Minafra I Cancemi P Marabeti MR Albanese NN Di Cara GTaormina P Marrazzo A Proteomic profiling of 13 paired duc-tal infiltrating breast carcinomas and non-tumoral adjacentcounterparts Proteomics Clinical Applications 2007 1118-129

29 Rowell C Carpenter DM Lamartiniere CA Chemoprevention ofbreast cancer proteomic discovery of genistein action in therat mammary gland J Nutr 2005 1352953S-2959S

30 Rucci N Recchia I Angelucci A Alamanou M Fattore AD FortunatiD Susa M Fabbro D Bologna M Teti A Inhibition of protein

kinase c-Src reduces the incidence of breast cancer metas-tases and increases survival in mice implications for therapyJ Pharmacol Exp Therapeutics (JPET) 2006 318161-172

31 Grover A Adamson ED Roles of extracellular matrix compo-nents in differentiating teratocarcinoma cells J Biol Chem1985 26012252-12258

32 Fuchshofer R Birke M Welge-Lussen U Kook D Luumltjen-Drecoll ETransforming growth factor-beta2 modulated extracellularmatrix component expression in cultured human opticnerve head astrocytes Invest Ophthalmol Vis Sci 2005 46568-578

33 Conesa A Gotz S Garcia-Gomez JM Terol J Talon M Robles MBlast2GO a universal tool for annotation visualization andanalysis in functional genomics research Bioinformatics 2005213674-3676

34 Altschul S Madden T Schaffer A Zhang J Zhang Z Miller W LipmanD Gapped BLAST and PSI-BLAST a new generation of pro-tein database search programs Nucleic Acids Res 1997253389-3402

35 Altschul S Gish W Miller W Myers E Lipman D Basic local align-ment search tool J Mol Biol 1990 21503-410

36 McEntyre J Ostell J (Eds) The NCBI handbook Bethesda (MD)National Library of Medicine (US) NCBI 2005

37 Ashburner M Ball CA Blake JA Botstein D Butler H Cherry JMDavis AP Dolinski K Dwight SS Eppig JT Harris MA Hill DP Issel-Tarver L Kasarskis A Lewis S Matese JC Richardson JE Ringwald MRubin GM Sherlock G Gene Ontology tool for the unificationof biology Nature Genetics 2000 2525-29

38 Ashburner M Lewis S On ontologies for biologists The GeneOntology ndash untangling the web Novartis Found Symp 2002 247

39 Davis GD Elisee C Newham DM Harrison RG New fusion pro-tein systems designed to give soluble expression inEscherichia coli Biotechnol Bioeng 1999 65382-388

40 Christendat D Yee A Dharamsi A Kluger Y Savchenko A Cort JRBooth V Mackereth CD Saridakis V Ekiel I Kozlov G Maxwell KLWu N McIntosh LP Gehring K Kennedy MA Davidson AR Pai EFGerstein M Edwards AM Arrowsmith CH Structural proteomicsof an archaeon Nat Struct Biol 2000 7903-909

41 Bertone P Kluger Y Lan N Zheng D Christendat D Yee A EdwardsAM Arrowsmith CH Montelione GT Gerstein M SPINE an inte-grated tracking database and data mining approach for iden-tifying feasible targets in high-throughput structuralproteomics Nucleic Acids Res 2001 292884-2898

42 Ignatoski KMW Verderame MF Lysis buffer composition dra-matically affects extraction of phosphotyrosine-containingproteins BioTechniques 1996 20794-796

43 Berkelman T Brubacher MG Chang H Important factors influ-encing protein solubility for 2-D electrophoresis BioRadiations2004 11431-32

44 Wang YP Wang XC Tian Q Yang Y Zhang Q Zhang JY Zhang YCWang ZF Wang Q Li H Wang JZ Endogenous overproductionof β-amyloid induces tau hyperphosphorylation anddecreases the solubility of tau in N2a cells Journal of NeuralTransmission 2006 1131723-1732

45 Baan B Dam Hv Zon GCM van der Maassen JA Ouwens DM TheRole of c-Jun N-terminal kinase p38 and extracellular sig-nal-regulated kinase in insulin-induced Thr69 and Thr71phosphorylation of activating transcription factor 2 MolecularEndocrinol 2006 201786-1795

46 Gao H Wu B Giese R Zhu Z Xom interacts with and stimu-lates transcriptional activity of LEF1TCFs implications forventral cell fate determination during vertebrate embryo-genesis Cell Research 2007 17345-356

47 DeSeau V Rosen N Bolen JB Analysis of pp60c-src tyrosinekinase activity and phosphotyrosyl phosphatase activity inhuman colon carcinoma and normal human colon mucosalcells J Cellular Biochem 1987 35113-128

48 Zapata JM Takahashi R Salvesen GS Reed JC Granzyme releaseand caspase activation in activated human T-lymphocytes JBiol Chem 1998 2736916-6920

49 Granier F Extraction of plant proteins for two-dimensionalelectrophoresis Electrophoresis 1988 1988 9712-718

50 Ames GF-L Nikaido K Two-dimensional gel electrophoresis ofmembrane proteins Biochemistry 1976 15616-623

51 OFarrell PH High resolution two-dimensional electrophore-sis of proteins J Biol Chem 1975 2504007-4021

Page 23 of 24(page number not for citation purposes)

Proteome Science 2008 630 httpwwwproteomescicomcontent6130

Publish with BioMed Central and every scientist can read your work free of charge

BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime

Sir Paul Nurse Cancer Research UK

Your research papers will be

available free of charge to the entire biomedical community

peer reviewed and published immediately upon acceptance

cited in PubMed and archived on PubMed Central

yours mdash you keep the copyright

Submit your manuscript herehttpwwwbiomedcentralcominfopublishing_advasp

BioMedcentral

52 Padgaonkar V Giblin F Reddy V Disulfide cross-linking of urea-insoluble proteins in rabbit lenses treated with hyperbaricoxygen Exp Eye Res 1989 49887-899

53 Harding JJ Conformational changes in human lens proteins incataract Biochem J 1972 12997-100

54 Fleschner CR Connexin 46 and connexin 50 in Selenite cata-ract Ophthalmic Research 2006 3824-28

55 Weber DJ McFadden PN A heterogenous set of urea-insolubleproteins in dividing PC12 pheochromocytoma cells is passedon to at least the generation of great-granddaughter cells JProtein Chem 1995 14283-289

56 Wu H Gao J Sharif WD Davidson MK Wahls WP Purificationfolding and characterization of Rec12 (Spo11) meioticrecombinase of fission yeast Protein Expression Purification 200438136-144

57 Moudjou M Paintrand M Vigues B Bornens M A human centro-somal protein is immunologically related to basal body-asso-ciated proteins from lower eukaryotes and is involved in thenucleation of microtubules J Cell Biol 1991 115129-140

Page 24 of 24(page number not for citation purposes)